diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml index f044897..df06152 100644 --- a/.github/workflows/ci.yml +++ b/.github/workflows/ci.yml @@ -96,8 +96,17 @@ jobs: - name: Setup Hugo uses: peaceiris/actions-hugo@v3 with: - hugo-version: 'latest' + hugo-version: '0.165.0' extended: true - - name: Build Hugo site - run: cd web && hugo --minify + - name: Build and test website + working-directory: web + run: npm run check + + - name: Validate rendered HTML + working-directory: web + run: npx --yes html-validate 'public/**/*.html' + + - name: Validate CSS + working-directory: web + run: npx --yes stylelint static/style.css diff --git a/.github/workflows/deploy-website.yml b/.github/workflows/deploy-website.yml index 9cc4751..5c2abde 100644 --- a/.github/workflows/deploy-website.yml +++ b/.github/workflows/deploy-website.yml @@ -18,8 +18,34 @@ concurrency: cancel-in-progress: false jobs: + verify: + name: Verify Website + runs-on: ubuntu-latest + steps: + - name: Checkout + uses: actions/checkout@v6 + + - name: Setup Hugo + uses: peaceiris/actions-hugo@v3 + with: + hugo-version: '0.165.0' + extended: true + + - name: Build and test website + working-directory: web + run: npm run check + + - name: Validate rendered HTML + working-directory: web + run: npx --yes html-validate 'public/**/*.html' + + - name: Validate CSS + working-directory: web + run: npx --yes stylelint static/style.css + deploy: name: Deploy to GitHub Pages + needs: verify runs-on: ubuntu-latest environment: name: github-pages @@ -32,11 +58,11 @@ jobs: - name: Setup Hugo uses: peaceiris/actions-hugo@v3 with: - hugo-version: 'latest' + hugo-version: '0.165.0' extended: true - name: Build Hugo site - run: cd web && hugo --minify + run: cd web && hugo --cleanDestinationDir --minify - name: Configure Pages uses: actions/configure-pages@v5 diff --git a/web/.gitignore b/web/.gitignore index ca204df..e4f6606 100644 --- a/web/.gitignore +++ b/web/.gitignore @@ -2,3 +2,6 @@ public/ resources/ .hugo_build.lock + +# Netlify CLI local site link (machine-specific, created by `netlify deploy`) +.netlify/ diff --git a/web/.htmlvalidate.json b/web/.htmlvalidate.json new file mode 100644 index 0000000..6e305a1 --- /dev/null +++ b/web/.htmlvalidate.json @@ -0,0 +1,7 @@ +{ + "extends": ["html-validate:recommended"], + "rules": { + "doctype-style": "off", + "no-trailing-whitespace": "off" + } +} diff --git a/web/content/_index.md b/web/content/_index.md index 726dc8e..3635b19 100644 --- a/web/content/_index.md +++ b/web/content/_index.md @@ -1,5 +1,5 @@ --- -title: "VocaMac - 100% Local Voice-to-Text for macOS" -description: "Free, 100% offline voice-to-text for macOS. Hold a hotkey, speak, and text appears at your cursor — private, fast, and powered by WhisperKit on Apple Silicon." -keywords: "voice to text macOS, offline dictation, speech to text mac, whisperkit, coreml speech recognition, push to talk mac, open source dictation, privacy voice typing, apple silicon voice recognition" +title: "VocaMac — native voice typing for macOS" +description: "Native voice typing for macOS. Hold a hotkey, speak, and text appears at your cursor — transcribed by a Whisper model running on your own Mac. Free and open source." +keywords: "voice to text macOS, on-device dictation, speech to text mac, whisperkit, coreml speech recognition, push to talk mac, open source dictation, privacy voice typing, apple silicon voice recognition" --- diff --git a/web/content/features/_index.md b/web/content/features/_index.md index 215ca71..c3b3057 100644 --- a/web/content/features/_index.md +++ b/web/content/features/_index.md @@ -1,5 +1,5 @@ --- title: "Features" -description: "Explore every VocaMac feature — push-to-talk dictation, 99+ languages, custom vocabulary, usage stats, and more. 100% local voice-to-text for macOS." +description: "Every VocaMac feature — push-to-talk dictation, language support, custom vocabulary, usage stats, and more. On-device voice typing for macOS." keywords: "vocamac features, voice to text features macOS, dictation app features, custom vocabulary, usage statistics, hotkey dictation" --- diff --git a/web/content/features/apple-silicon-native.md b/web/content/features/apple-silicon-native.md index 4707db1..91cb5e7 100644 --- a/web/content/features/apple-silicon-native.md +++ b/web/content/features/apple-silicon-native.md @@ -1,53 +1,33 @@ --- title: "Apple Silicon Native" -subtitle: "CoreML + Metal + Neural Engine acceleration via WhisperKit. Blazing fast on M1/M2/M3/M4." -description: "VocaMac runs natively on Apple Silicon using CoreML and the Neural Engine for hardware-accelerated speech recognition. No cloud, no CPU bottleneck." -keywords: "apple silicon speech recognition, coreml voice to text, neural engine dictation, whisperkit macOS, M1 M2 M3 M4 voice typing, hardware accelerated transcription mac" +subtitle: "A native macOS app with local WhisperKit/CoreML transcription on Apple Silicon." +description: "VocaMac's stable release is built for Apple Silicon Macs and runs its WhisperKit speech-processing path locally after the model is available." +keywords: "apple silicon speech recognition, coreml voice to text, whisperkit macOS, local dictation mac, hardware accelerated transcription mac" icon: "⚡" --- -## Designed for Modern Mac Hardware +## Built for Apple Silicon -VocaMac is built from the ground up for Apple Silicon Macs. Using CoreML and WhisperKit, the app harnesses the dedicated Neural Engine in your M1, M2, M3, or M4 chip for speech recognition that's faster and more efficient than anything running on the CPU alone. - -This isn't just optimization. It's a fundamental architectural advantage. While cloud-based dictation apps send your voice across the internet and wait for responses, VocaMac processes everything locally on your Mac in milliseconds. - -## Neural Engine Acceleration +VocaMac's stable release targets Apple Silicon Macs running macOS 13 Ventura or later. The app is a native SwiftUI menu-bar client, with WhisperKit and CoreML providing the stable release's speech-processing path. ![VocaMac Settings showing model management on Apple Silicon](/screenshots/settings-models.png) -Apple Silicon's Neural Engine is a specialized processor designed for machine learning tasks. When VocaMac transcribes your voice, it offloads the heavy computational work to this dedicated hardware. - -The result is remarkable. Transcription happens in real time. You finish speaking, and your words appear before you've had time to take your next breath. This responsiveness creates a natural, uninterrupted flow that makes dictation feel less like using an app and more like a natural extension of how you type. - -The Neural Engine operates independently from your Mac's CPU and GPU. This means transcription doesn't compete with your other work. You can continue editing documents, reviewing emails, or running complex applications while VocaMac transcribes in the background without any performance impact. - -## CoreML and Metal Integration +The exact speed and memory profile depends on the selected model, recording length, available memory, and what else your Mac is doing. VocaMac exposes model sizes and resource guidance in Settings so you can choose an appropriate trade-off. -CoreML is Apple's framework for on-device machine learning. VocaMac uses CoreML to run the Whisper speech recognition model locally, with zero reliance on cloud services. Your voice never leaves your Mac. +## Local Processing Boundary -Metal, Apple's graphics API, provides additional acceleration for certain computational tasks. Together, CoreML and Metal ensure maximum efficiency while keeping your data private and your Mac responsive. +After the selected model is downloaded, the Whisper processing path runs on your Mac. Dictation audio is not sent to a Voca cloud endpoint. The first model download, release downloads, and update checks are separate network actions. -WhisperKit, the framework powering VocaMac, was specifically engineered to take full advantage of Apple Silicon. It automatically detects your hardware and uses the optimal execution path. CoreML for neural processing. Metal for graphics-related computation. The CPU for coordination. All working in concert. +This boundary makes VocaMac useful when you want local speech processing or need to work with sensitive notes. It does not make transcription instantaneous or set a fixed accuracy or latency across every Mac. -## Unmatched Performance +## Model Choices -Benchmark after benchmark shows the same pattern. On an M1 Mac, medium-sized Whisper models transcribe audio roughly 3 to 5 times faster than real time. A 30-second audio clip finishes transcribing in 6 to 10 seconds. CPU-only inference on machines without a Neural Engine can take 90 seconds or more for the same task — which is exactly why VocaMac ships for Apple Silicon only. +The stable catalog includes Tiny, Base, Small, compact Large v3, Distil Large, and Large v3 variants. Smaller models generally need less disk space and memory; larger models may improve accuracy while taking longer to download and process. -This performance gap only widens with larger, more accurate Whisper models. The base model runs instantly. The small model takes a few seconds. The medium model, which offers near-human accuracy, completes in under a minute on Apple Silicon. Cloud-based services may offer similar speed, but they require internet, demand subscription fees, and raise privacy concerns. - -## Battery Efficiency and Thermal Design - -Because the Neural Engine is purpose-built for machine learning, it's remarkably power efficient. A task that would consume significant battery power if handled by the CPU uses a fraction of the energy when offloaded to the Neural Engine. - -This efficiency translates to cooler operation. Your Mac's fan rarely needs to spin up. The app runs silently, without adding heat or noise to your workspace. - -On M-series Macs, you can use VocaMac all day without noticing any battery drain. The app is designed to be always available, always responsive, and always respectful of your hardware's resources. +Start with a smaller model if you are checking the workflow or have limited disk space. Move up the catalog when accuracy matters and your Mac has enough headroom. The Settings screen shows the current download state and lets you switch the selected model. ## Why It Matters -Apple Silicon isn't just faster. It represents a different philosophy. Apple designed these chips to handle the tasks people actually care about with remarkable efficiency. - -For dictation, this philosophy makes a difference. Faster transcription means less waiting. Local processing means your voice stays private. Efficient hardware means your Mac stays cool and responsive. Taken together, these advantages make VocaMac the most natural voice-to-text experience on Mac. +Apple Silicon gives the native app a consistent local hardware target and lets CoreML use Apple's on-device execution stack. You keep control over model downloads, microphone permissions, Accessibility access, and the text insertion destination. -Whether you have an M1, M2, M3, or M4 Mac, you're getting the full benefit of your hardware's capabilities. VocaMac is native. It's fast. It's efficient. And it's built specifically for your Mac. +For the stable release's complete version, download links, supported language hints, and permissions, see the [installation guide](/#install) and [model catalog](/#models). For newer engines and experiments on `main`, see the [nightly boundary](/#models). diff --git a/web/content/features/custom-vocabulary.md b/web/content/features/custom-vocabulary.md index 3c05cef..f455212 100644 --- a/web/content/features/custom-vocabulary.md +++ b/web/content/features/custom-vocabulary.md @@ -1,7 +1,7 @@ --- title: "Custom Vocabulary" subtitle: "Teach VocaMac the names, brands, and jargon you use so they're transcribed right, every time." -description: "Add names, technical terms, and jargon to VocaMac's custom vocabulary so Whisper spells them correctly. Processed 100% locally, like everything else in VocaMac." +description: "Add names, technical terms, and jargon to VocaMac's custom vocabulary so the Whisper path can use your preferred spellings locally." keywords: "custom vocabulary dictation, whisper glossary, proper noun transcription, technical jargon speech to text, fix misspelled names dictation macOS" icon: "📝" --- @@ -23,7 +23,7 @@ Open **Settings → General → Custom Vocabulary** and type the terms you care kubectl, PostgreSQL, nginx, Grafana ``` -VocaMac passes these to Whisper as a transcription hint, biasing it toward your spelling. Keep the list focused: the model uses roughly the first 50–100 words, so put your most-mistranscribed terms at the top. There's no training step and no delay. Add a term, and your very next dictation already knows it. +VocaMac passes these to Whisper as a transcription hint, biasing it toward your spelling. Keep the list focused and put your most-mistranscribed terms near the top; WhisperKit applies its own prompt-token budget. There is no training step. The updated list is used on the next transcription attempt, although it remains a hint rather than a promise. For best results, enter terms in the language you dictate and set a matching transcription language above. In Auto-detect, your vocabulary can even nudge VocaMac toward the right language. diff --git a/web/content/features/double-tap-toggle.md b/web/content/features/double-tap-toggle.md index 1fd41f3..0a66fec 100644 --- a/web/content/features/double-tap-toggle.md +++ b/web/content/features/double-tap-toggle.md @@ -22,7 +22,7 @@ Configure your hotkey and set VocaMac to Toggle mode. When recording begins: 4. Speak your message, question, email, code comment, or anything else 5. Double-tap your hotkey again to stop recording and transcribe -The transcription happens instantly. Your text appears at your cursor position. If you double-tap again, VocaMac starts a new recording session. +After you stop, VocaMac processes the recording and inserts the result at your cursor. Processing time varies with recording length, selected model, and Mac hardware. ## Configurable Double-Tap Threshold diff --git a/web/content/features/fully-configurable.md b/web/content/features/fully-configurable.md index 94de824..efe7b00 100644 --- a/web/content/features/fully-configurable.md +++ b/web/content/features/fully-configurable.md @@ -20,23 +20,23 @@ Your hotkey is how you activate VocaMac. The app respects how you work and lets Start from a preset — Right Option, Left Option, Right Command, Right Shift, Right Control, Fn, or a function key (F5–F12). Prefer something else? Click **Record**, press any key, and VocaMac captures it (press Escape to cancel). Your choice shows up as a "Custom" key, and while VocaMac is running that key is reserved for activation. -The hotkey works from anywhere on your Mac. Launch VocaMac from Mail, from Slack, from a web browser, or from any application. As long as VocaMac is running, your hotkey is active and ready. +The hotkey is global while VocaMac is running, so you can trigger it while working in Mail, Slack, a browser, or another app. The target app still needs to expose a text field for insertion. ## Model Management -VocaMac includes multiple Whisper models, each offering different tradeoffs between speed and accuracy. The settings window gives you full control over which models are available on your Mac. +VocaMac includes multiple Whisper models, each offering different tradeoffs between speed, memory, and accuracy. The settings window gives you control over which models are available on your Mac. -The base model is tiny (just 140 megabytes) and transcribes nearly as fast as you can speak. It's perfect for quick notes or when speed matters more than perfect accuracy. +The Base model is a small download and transcribes quickly. It is useful for quick notes or when speed and disk space matter more than maximum accuracy. -The small and medium models offer progressively better accuracy at the cost of slightly longer transcription times. They're ideal for important documents, interviews, or situations where every word needs to be perfect. +Small, compact Large, and Distil variants offer different accuracy and speed trade-offs. Choose among the models shown in Settings rather than assuming every legacy Whisper size is part of the stable catalog. -The model management interface shows how much disk space each model uses. Download the models you want. Delete the ones you don't need. Switch between models instantly. You're always in control of what's installed on your Mac. +The model management interface shows how much disk space each model uses. Download the models you want and switch the selected model from Settings. Model deletion is available in newer nightly/source builds, not the v0.7.2 stable catalog. ## Audio Settings Audio behavior can make or break the dictation experience. VocaMac includes granular controls for how it handles microphone input and silence detection. -Set the silence detection threshold to match your environment. In a quiet office, a sensitive threshold works well. In a noisier space, you might want to increase it to avoid accidental transcriptions. Adjust the threshold, speak a few test words, and instantly see how the app responds. +Set the silence detection threshold to match your environment. In a quiet office, a lower threshold may work well; in a noisier space, a higher threshold can help avoid accidental stops. Test the setting with your microphone and speaking style. Configure the maximum recording duration. If you prefer shorter bursts of dictation, set a reasonable limit. The app will automatically stop recording when you reach it, preventing accidental marathon recording sessions. diff --git a/web/content/features/languages.md b/web/content/features/languages.md index 46587d9..cace4a2 100644 --- a/web/content/features/languages.md +++ b/web/content/features/languages.md @@ -1,75 +1,31 @@ --- -title: "99+ Languages" -subtitle: "Auto-detect or specify. English, Spanish, French, German, Chinese, Japanese, and many more." -description: "VocaMac supports 99+ languages for speech recognition powered by OpenAI's Whisper model. Auto-detect the spoken language or set it manually for best accuracy." -keywords: "multilingual dictation macOS, whisper language support, speech to text languages, auto detect language voice, multilingual voice typing mac, foreign language dictation" +title: "Language support" +subtitle: "Let Whisper detect a language or choose one of the languages exposed in VocaMac's stable settings picker." +description: "VocaMac's stable Whisper path supports automatic detection and 17 selectable language hints, with all transcription running locally on your Mac." +keywords: "multilingual dictation macOS, whisper language support, speech to text languages, auto detect language voice, multilingual voice typing mac" icon: "🌍" --- -## Speak in Your Language +## Speak in your language ![VocaMac Settings showing language configuration](/screenshots/settings-general.png) -VocaMac supports over 99 languages. Whether you dictate in English, Spanish, Mandarin, Arabic, or any of dozens of other languages, the app understands you. More importantly, it transcribes with high accuracy regardless of which language you're using. +VocaMac's stable Whisper path supports automatic detection plus 17 language hints in Settings: English, Spanish, French, Italian, German, Portuguese, Dutch, Chinese, Japanese, Korean, Hindi, Arabic, Russian, Turkish, Polish, Swedish, and Ukrainian. -This capability comes from OpenAI's Whisper model, one of the most advanced speech recognition systems ever created. Whisper was trained on 680,000 hours of multilingual audio data. It understands not just major languages, but regional dialects, accents, and linguistic variations. +The underlying Whisper model was trained on more languages than VocaMac exposes as manual settings. The website describes the picker users can actually use, not a larger training-data figure. -For people who work across languages, this is transformative. No more switching between different apps for different languages. No more cloud services that only support a handful of languages. VocaMac handles all of them with a single, elegant interface. +## Automatic or chosen -## Auto-Detection or Manual Selection +Automatic detection is useful when you want VocaMac to infer the language from a full phrase. A manual hint can help when an utterance is short, an accent is unfamiliar, or you are dictating a predictable language for a long session. -VocaMac offers two approaches to language handling. You can let the app automatically detect which language you're speaking, or you can manually specify a language for consistent accuracy. +Open **Settings → General → Transcription Language** to change the hint. The selected language is saved locally and can be changed without an account or a cloud service. -Auto-detection is powerful for multilingual environments. If you're bilingual or trilingual, you might mix languages within the same conversation. VocaMac listens, analyzes what you're saying, and transcribes accurately without any configuration. Speak English one moment, switch to Spanish the next, and the app follows along naturally. +## Model size still matters -For best results with auto-detection, speak complete phrases or sentences. Single words or very short utterances might occasionally confuse the language detector. But in normal conversation, auto-detection is remarkably accurate. +Every Whisper model has different speed, memory, and accuracy trade-offs. Smaller models are useful for quick notes; larger models generally give the recognizer more room for difficult audio and non-English speech. The model table on the home page reflects the stable release's actual catalog. -Manual language selection is ideal when you know you'll be dictating in a specific language. Open settings, choose your language, and enjoy transcription optimized for that particular language. The app uses this information to improve accuracy and context understanding. +## Local processing -Switching languages is instant. No restarts. No downloads. Just open settings, select a different language, and start speaking. +Language choice does not change the processing boundary: the selected Whisper model runs on your Mac. The model may need a one-time download, but VocaMac does not send dictation audio to a Voca cloud endpoint. -## Supported Languages - -The full list of supported languages includes: - -English, Spanish, French, German, Italian, Portuguese, Russian, Chinese (Mandarin and Cantonese), Japanese, Korean, Arabic, Hebrew, Turkish, Dutch, Greek, Polish, Romanian, Swedish, Finnish, Danish, Norwegian, Czech, Hungarian, Slovak, Bulgarian, Croatian, Serbian, Slovenian, Lithuanian, Latvian, Estonian, Thai, Vietnamese, Indonesian, Filipino, Afrikaans, and many more. - -Major world languages are represented. Regional languages are included. Less commonly spoken languages have support too. Whether you're dictating in a major commercial language or a less common one, Whisper can handle it. - -## Accuracy Across Languages - -Whisper's training data ensures consistent accuracy across languages. English, being the most represented language in the training set, achieves the highest accuracy. But Spanish, French, German, Chinese, and Japanese all achieve excellent results. - -Smaller languages have slightly lower accuracy, but still remain highly usable. The base Whisper model performs admirably even on less commonly supported languages. Larger models (small, medium, large) improve accuracy further, especially for non-English languages. - -Regional accents are handled well. Whether you speak British English, Australian English, or American English, the transcription is accurate. The same applies to Spanish speakers from Spain, Latin America, or elsewhere. Whisper recognizes regional variations without requiring manual configuration. - -## Model Performance by Language - -All Whisper model sizes (base, small, medium, large) support all 99+ languages. The choice of which model to use depends on your accuracy needs and hardware. - -The base model, despite its size, handles most languages well. For casual dictation in any language, it's sufficient. - -The small model noticeably improves accuracy, especially for non-English languages. If you frequently dictate in Spanish, French, German, or Asian languages, the small model is a worthwhile upgrade. - -The medium model achieves near-human accuracy and is particularly strong with non-English languages. It's the sweet spot for serious, multilingual work. - -The large model is the most accurate but requires more processing time. For languages where perfection matters (legal documents, medical transcription, academic writing), it's worth the extra seconds of processing. - -## Real-World Multilingual Workflow - -Imagine you're a translator working on Spanish documents. You open VocaMac, set the language to Spanish, and begin dictating. The app transcribes with accuracy that rivals human typists. - -An hour later, you switch projects. Now you're reviewing French content. Change the language setting to French, and you're ready to dictate again. The transition takes seconds. - -Or picture yourself working internationally. Your team uses English, but you're collaborating with colleagues who prefer working in their native languages. Enable auto-detection and speak naturally. VocaMac adapts to whoever is dictating. - -These scenarios that would be cumbersome with traditional dictation tools feel natural with VocaMac. You don't think about language switching. You just speak, and the app keeps up. - -## Privacy in Any Language - -Regardless of which language you choose, your voice never leaves your Mac. VocaMac processes everything locally. Whether you're dictating in English, Mandarin, Arabic, or any other language, complete privacy is guaranteed. - -This is a critical distinction from cloud-based services. Those systems must send your voice to remote servers to perform transcription, even if those servers are in other countries. With VocaMac, your voice stays on your device, transcribed by software running in your own Mac. - -VocaMac brings professional-quality, multilingual dictation to your Mac while respecting your privacy and your choice of language. Speak your language. VocaMac listens. +Accuracy depends on the model, microphone, accent, background noise, and length of the utterance. Treat the transcript as editable text, especially for names, specialist vocabulary, and short phrases. diff --git a/web/content/features/launch-at-login.md b/web/content/features/launch-at-login.md index 3732dc7..c30a67d 100644 --- a/web/content/features/launch-at-login.md +++ b/web/content/features/launch-at-login.md @@ -1,110 +1,21 @@ --- -title: "Launch at Login" -subtitle: "Start VocaMac automatically when you log in. Always ready when you need it." -description: "VocaMac can start automatically when you log into your Mac using the modern SMAppService API. Always ready in your menu bar without manual launching." -keywords: "launch at login macOS, auto start menu bar app, SMAppService, startup app mac, always ready dictation, login item macOS app" +title: "Launch at login" +subtitle: "Start VocaMac with your Mac so the menu-bar shortcut is ready when you are." +description: "VocaMac can register as a macOS login item using SMAppService and remain ready in the menu bar without a manual launch." +keywords: "launch at login macOS, auto start menu bar app, SMAppService, startup app mac, always ready dictation" icon: "🚀" --- -## How It Works - -Launch at Login is a simple but powerful feature: VocaMac starts automatically when you log into your Mac. You'll never have to manually launch the app. It's always there in your menu bar, ready to dictate. - -VocaMac uses macOS's modern **SMAppService API** to register itself as a login item. This is Apple's recommended way to add apps to the login sequence, and it works seamlessly with System Settings on macOS 13 and later. - -When you enable Launch at Login in VocaMac's settings, the app registers itself with the system. The next time you log in (or restart your Mac), VocaMac will launch automatically and appear in your menu bar. No dialogue boxes. No extra windows. It simply starts and waits for you to dictate. - -## Enabling Launch at Login +## Ready in the menu bar ![VocaMac Settings showing Launch at Login toggle](/screenshots/settings-general.png) -There are two ways to enable this feature: - -**Via Settings** - -1. Open VocaMac and click the menu bar icon -2. Select **Settings** from the popover -3. Go to **General → Launch at Login** -4. Toggle the switch to enable - -**Via Setup Wizard** - -If you're setting up VocaMac for the first time, the Setup Wizard will offer to enable Launch at Login. Just check the box during initial setup. - -Once enabled, the app will start automatically on your next login or restart. - -## Sync with System Settings - -VocaMac's Launch at Login toggle is always in sync with macOS System Settings. If you enable Launch at Login in VocaMac, it will also appear in: - -**System Settings → General → Login Items** - -You can manage it from either place. If you disable it in System Settings, VocaMac's toggle will reflect that change. This bidirectional sync ensures you always have a single source of truth. - -You can verify this yourself: - -1. Enable Launch at Login in VocaMac -2. Open **System Settings → General → Login Items** -3. Scroll to the "Allow in the Login Items" section -4. VocaMac will be listed there - -If you ever want to disable it, you can do so from System Settings without opening VocaMac. - -## Always Available in Your Menu Bar - -With Launch at Login enabled, VocaMac is always available. There's no startup delay. The moment you log in, the app is ready: - -- The menu bar icon appears instantly (along with your other menu bar apps) -- All your settings are loaded (activation mode, hotkey, silence thresholds, etc.) -- Accessibility permissions are already granted (from your previous session) -- The WhisperKit model is cached, so transcription is fast - -This means you can start dictating immediately. No waiting for the app to launch. No loading screens. Just open any text field and press your hotkey. - -## Lightweight Resource Usage - -VocaMac is designed to be a background app. It uses minimal system resources when idle: - -- **Memory**: approximately 50-100 MB at rest -- **CPU**: near zero when not recording -- **Battery**: negligible impact on battery life -- **Network**: no network activity unless you're downloading a new model - -With Launch at Login enabled, VocaMac won't slow down your Mac's startup time or consume meaningful resources while you work. It sits quietly in the background until you need it. - -## Why Use Launch at Login? - -Enabling this feature has several benefits: - -**Convenience**: no need to manually launch VocaMac after every restart or login - -**Reliability**: you won't accidentally forget to launch the app and lose dictation access mid-session - -**Consistency**: all your settings are preserved across restarts - -**Simplicity**: one less thing to think about; VocaMac just works - -**Professional workflow**: for users who rely on voice dictation throughout the day, having it always available is essential - -## Disabling Launch at Login - -If you ever decide you don't want VocaMac to start automatically, you can disable it anytime: - -1. Open VocaMac settings -2. Go to **General → Launch at Login** -3. Toggle the switch to disable - -Or disable it directly in **System Settings → General → Login Items → Allow in the Login Items** section. - -VocaMac will no longer start at login, but it will still be in your Applications folder and ready to launch manually whenever you want. +Enable **Launch at Login** in VocaMac's General settings and macOS starts the app when you sign in. VocaMac uses Apple's SMAppService login-item API, and the setting is also visible in **System Settings → General → Login Items**. -## Technical Details +This controls when the app starts; it does not force a speech model to stay loaded. Model loading and model keep-alive behavior remain separate settings so you can choose the resource trade-off that fits your Mac. -VocaMac uses the **SMAppService** API from the ServiceManagement framework, which is the modern, recommended approach for login items on macOS 13 and later. This API is: +## Local and quiet -- **More reliable** than older helper app approaches -- **Transparent** in System Settings -- **Secure** since it doesn't require elevated privileges -- **Reversible** at any time with a simple toggle +The menu-bar app can be ready without opening a large window. Audio is captured only during a recording session, and transcription uses the selected local model. Model downloads, update checks, and release downloads are separate network actions; enabling launch at login does not turn on website analytics or telemetry. -If you're curious about how this works under the hood, VocaMac's source code is open and available on GitHub. The implementation is straightforward and uses no workarounds or hacks. +You can disable the login item from VocaMac or from macOS System Settings. Release builds are Developer ID signed; ad-hoc source builds may require permissions again after rebuilding. diff --git a/web/content/features/push-to-talk.md b/web/content/features/push-to-talk.md index 3df934f..f0923e6 100644 --- a/web/content/features/push-to-talk.md +++ b/web/content/features/push-to-talk.md @@ -15,7 +15,7 @@ Push-to-Talk is the default activation mode in VocaMac. It's designed to feel as 3. **Release** the key when you're done 4. **Text appears** at your cursor, wherever you're typing -The entire flow takes less than a second from release to text appearing on screen. There's no delay, no loading spinner, no waiting. Just your words, typed out. +After you release the key, VocaMac processes the recording and inserts the result at the active cursor. The time varies with recording length, selected model, and Mac hardware. ## Why Push-to-Talk? @@ -25,7 +25,7 @@ Push-to-Talk eliminates that entirely: - **No ambiguity**: if your finger is on the key, you're recording. If it's not, you're not. - **No forgotten recordings**: release the key and recording stops automatically. -- **Instant feedback**: the menu bar icon turns green the moment you press, and returns to normal when you release. +- **Clear feedback**: the menu bar and popover expose the active recording state; color is not the only signal. - **Muscle memory**: after a few uses, it becomes second nature. Like holding Shift to capitalize. ## Choosing Your Hotkey @@ -50,13 +50,13 @@ You can change your hotkey anytime in **Settings → General → Activation Key* While recording, VocaMac gives you clear visual cues so you always know what's happening: -- **Menu bar icon** turns green to indicate active recording +- **Menu bar icon and status text** indicate active recording - **Audio level indicator** shows real-time input volume in the popover - **Cursor indicator** (optional) shows a floating mic icon near your text cursor -## Works Everywhere +## Works Across Text Fields -Push-to-Talk works in any application that accepts text input: +Push-to-Talk is intended for applications and fields that accept macOS text insertion, including: - Text editors and IDEs (VS Code, Xcode, Sublime Text) - Browsers (Chrome, Safari, Firefox, Arc) @@ -65,7 +65,7 @@ Push-to-Talk works in any application that accepts text input: - Terminal emulators - Email clients -VocaMac injects text at your cursor position using macOS accessibility APIs, so it works system-wide without any app-specific integrations. +VocaMac injects text at your cursor position using macOS accessibility APIs. The exact result depends on the target app, the active field, and Accessibility permission. ## Compared to Double-Tap Toggle diff --git a/web/content/features/silence-detection.md b/web/content/features/silence-detection.md index 051fef7..2edf42c 100644 --- a/web/content/features/silence-detection.md +++ b/web/content/features/silence-detection.md @@ -28,7 +28,7 @@ Every environment is different. A quiet home office has lower background noise t This controls how sensitive the detector is to quiet sounds. A lower threshold is more aggressive (stops recording sooner), while a higher threshold is more lenient (requires more silence before stopping). - **Low threshold (20-40%)**: use in quiet environments where you want quick stops -- **Medium threshold (50-60%)**: the default sweet spot for most users +- **Medium threshold (50-60%)**: a useful starting point for many environments - **High threshold (70-90%)**: use in noisy environments to prevent accidental cutoffs **Silence Duration (0.5-3 seconds)** @@ -63,7 +63,7 @@ One concern with any silence detection system is false positives: stopping recor **Minimum recording duration**: VocaMac won't stop recording in the first 0.3 seconds, even if silence is detected, to avoid cutting off initial words -**Adaptive thresholds**: the silence detector continuously adapts to the ambient noise floor in your environment, so sudden sounds (a door closing, a car horn) won't trigger a false stop +**Audio-level feedback**: the popover exposes the current input level so you can tune the threshold against your actual microphone and room **Real-time monitoring**: you can see the audio level indicator in the VocaMac popover in real time, so you know exactly how loud the detector thinks your speech is @@ -71,13 +71,13 @@ One concern with any silence detection system is false positives: stopping recor VocaMac recently fixed a critical silence detection bug: in rare cases, audio at the very beginning of a recording could be lost if the silence detector kicked in too aggressively during the initial setup phase. This has been corrected. -Now, VocaMac guarantees that: +The current implementation is designed to: 1. No audio is lost at the start of recording 2. The minimum recording duration (0.3 seconds) is respected -3. The silence threshold always accounts for the detected noise floor +3. Keep silence detection scoped to the configured threshold and duration -This means you can trust silence detection completely. Your words won't be cut off, and nothing will be lost. +As with any energy-based detector, test the threshold and duration with your microphone and speaking style before relying on it for long dictations. ## Tips for Best Results diff --git a/web/content/features/smart-model-selection.md b/web/content/features/smart-model-selection.md index 908bd40..40bebe9 100644 --- a/web/content/features/smart-model-selection.md +++ b/web/content/features/smart-model-selection.md @@ -1,73 +1,31 @@ --- -title: "Smart Model Selection" -subtitle: "Auto-detects your Apple Silicon hardware and RAM, then recommends the optimal Whisper model." -description: "VocaMac automatically detects your Mac hardware and recommends the best Whisper model for your system. Choose from Tiny to Large v3 based on your needs." +title: "Smart model selection" +subtitle: "VocaMac reads your Apple Silicon hardware and helps you choose a Whisper model that fits its memory and accuracy trade-offs." +description: "VocaMac recommends a Whisper model based on your Mac's hardware, while keeping model choice and downloads under your control." keywords: "whisper model selection, auto detect hardware macOS, apple silicon whisper, coreml speech model, best whisper model mac, RAM based model recommendation" icon: "🧠" --- -## Intelligent Hardware Detection +## A recommendation, not a lock-in -Every Apple Silicon Mac is different. An 8 GB MacBook Air has very different headroom than a 32 GB Mac Studio. VocaMac detects your exact hardware configuration and recommends the Whisper model that delivers the best balance of accuracy and speed on your specific machine. +VocaMac can inspect the Mac's Apple Silicon hardware and available memory to recommend a practical model. You can keep that recommendation, choose another model, or switch later from Settings. -On first launch, VocaMac analyzes your chip generation, CPU core count, Neural Engine availability, and installed RAM. It then suggests the optimal model tier. You remain free to choose any model, but the recommendation gets you great results immediately without guesswork. +> **Note:** The stable release is an Apple Silicon build. Intel Macs are not supported. -> **Note:** VocaMac is Apple Silicon only — it does not run on Intel Macs. +## The stable model catalog -## Five Model Tiers +![VocaMac Settings showing model management and system information](/screenshots/settings-models.png) -![VocaMac Settings showing model management and system info](/screenshots/settings-models.png) +The stable release includes Tiny, Base, Small, compact Large v3 and Distil Large variants, and the full Large v3 model. Their approximate download sizes range from 39 MB to 3.1 GB. VocaMac shows the local model state and resource guidance in Settings. -VocaMac supports five Whisper model sizes, from lightweight to highly accurate. Each model is optimized for CoreML and runs entirely on your Mac. +Start with Small on an 8 GB Mac if you want a balanced default. Choose Tiny or Base for a smaller download and lower memory use. Choose a compact Large or Distil model when accuracy matters and your Mac has the headroom. The full Large v3 model needs substantially more disk space and memory. -**Tiny (39 MB)** -The fastest option with minimal memory overhead. Suitable for Macs with 4-8GB RAM or when you prioritize speed over perfect accuracy. Transcription completes in real-time or faster. Accuracy drops slightly compared to larger models, but remains acceptable for casual note-taking and quick messages. +## Download, verify, switch -**Base (140 MB)** -A solid middle ground. Runs efficiently on any Mac with 8GB RAM or more. Offers noticeably better accuracy than Tiny while maintaining very fast transcription speeds. This is often the recommended model for most users. +The first use of a model can download its files from the documented WhisperKit model repository. VocaMac caches the model locally, verifies the downloaded assets, and uses the selected model for later recordings. You can keep more than one model; model deletion is part of the newer nightly/source lane rather than the v0.7.2 stable catalog. -**Small (465 MB)** -For users with 16GB RAM seeking higher accuracy. Transcription speeds remain fast on modern Macs. Accuracy improves substantially. Recommended for professional writing, coding, and applications where precision matters. +The model choice affects speed, memory, and accuracy; it does not change the privacy boundary. Transcription remains on-device after the selected model is available. -**Medium (1.5 GB)** -High accuracy for demanding use cases. Requires 16GB RAM or more. Transcription remains reasonably fast on Apple Silicon Macs. Excellent for technical documentation, medical transcription, and content creation where every word counts. +## CoreML on Apple Silicon -**Large v3 (3 GB)** -The most accurate model available. Peak performance on Apple Silicon Macs with 32GB or more RAM. Transcription speed may reach 2-3 seconds per minute of audio. Use when maximum accuracy is essential and speed is secondary. - -## Hardware-Based Recommendations - -VocaMac provides tailored suggestions: - -- **MacBook Air (M1/M2, 8 GB)**: Base or Small model recommended -- **MacBook Pro (M1/M2/M3, 16 GB)**: Small or Medium model recommended -- **Mac mini (M2, 16 GB)**: Small or Medium model recommended -- **Mac Studio (M2/M3 Max, 32 GB+)**: Medium or Large v3 model recommended - -These recommendations balance your hardware capabilities with practical transcription speeds. Your actual choice depends on your accuracy requirements and tolerance for processing time. - -## Downloading and Switching Models - -Models download on-demand. First use of a model triggers a download (requires internet connection). Subsequent launches use the cached model. Each model includes a checksum verification to ensure integrity. - -Switch models instantly in VocaMac settings. No restart required. Your next recording uses the newly selected model. You can maintain multiple models on disk and switch between them based on context. Recording a technical meeting? Switch to Small. Quick email dictation? Use Tiny. - -Model storage uses your local disk space. VocaMac shows disk usage for each model. Delete unused models to reclaim space. - -## CoreML Optimization - -All Whisper models in VocaMac are converted to Apple's CoreML format. This optimization ensures: - -- Native execution on Apple Silicon using the Neural Engine -- GPU offload through Metal Performance Shaders for compute-heavy ops -- Minimal energy consumption (important for battery life on laptops) -- No cloud dependencies or external API calls -- Complete privacy (all processing happens locally) - -CoreML models leverage your Mac's specialized ML hardware. Apple Silicon Macs see dramatic speed improvements compared to generic ML frameworks. You get the best possible performance from your hardware. - -## Choose Your Balance - -VocaMac trusts you to make the final choice. The automatic recommendation helps you start immediately, but you control which model runs on your Mac. Prioritize speed on some machines. Prioritize accuracy on others. Change your preference whenever your needs evolve. - -Your dictation workflow adapts to your hardware and preferences. That is the VocaMac approach. +WhisperKit supplies the CoreML model path used by the stable release. VocaMac coordinates the model on your Mac instead of sending audio to a hosted speech API. Actual speed varies by chip, memory pressure, recording length, and model size, so the site avoids unsupported benchmark promises. diff --git a/web/content/features/system-wide-text-injection.md b/web/content/features/system-wide-text-injection.md index 7a80134..44100e2 100644 --- a/web/content/features/system-wide-text-injection.md +++ b/web/content/features/system-wide-text-injection.md @@ -1,62 +1,27 @@ --- -title: "System-Wide Text Injection" -subtitle: "Transcribed text appears wherever your cursor is. Browsers, Slack, VS Code, spreadsheets, terminals. Everywhere." -description: "VocaMac injects transcribed text directly at your cursor position in any macOS application using accessibility APIs. No copy-paste needed." +title: "System-wide text insertion" +subtitle: "Put the transcript at the active cursor in the apps where macOS permits accessibility-driven text insertion." +description: "VocaMac uses macOS accessibility APIs and a clipboard-safe insertion path to place local transcripts at the active cursor." keywords: "system wide dictation macOS, voice typing any app, text injection accessibility, dictation VS Code, voice to text terminal, speech to text everywhere mac" icon: "⌨️" --- -## Dictation Without Limits - -Traditional voice-to-text on macOS requires you to speak, copy the result, then paste it into your app. VocaMac skips that friction entirely. When you finish dictating, your words appear instantly at your cursor, regardless of which application you're using. - -This works in your browser, code editor, terminal, spreadsheet, or any other macOS app. No special setup per application. No compatibility lists to check. Just speak and type. - -## How It Works +## Dictate where you already work ![VocaMac popover showing transcription result](/screenshots/popover-panel.png) -VocaMac uses macOS accessibility APIs to detect your cursor position in the foreground application, then injects text directly into the input field. This is the same technology that powers accessibility features like VoiceOver and Switch Control. - -When you stop recording, VocaMac: - -1. Retrieves the cursor position from the active application -2. Sends your transcribed text character by character via the accessibility layer -3. Places the text exactly where your cursor is - -The process takes just milliseconds. There is no intermediate clipboard step, which means your existing clipboard contents remain untouched. - -## Supported Applications - -System-wide text injection works with any macOS application that respects standard input methods. This includes: - -- **Browsers**: Gmail, Google Docs, Notion, web-based IDEs, comment sections -- **Code Editors**: VS Code, Xcode, Sublime Text, Nova, BBEdit, Terminal -- **Communication**: Slack, Discord, Twitter, Mastodon, Signal, iMessage -- **Productivity**: Microsoft Word, Google Sheets, Numbers, Notion, Apple Pages -- **Developer Tools**: iTerm2, Zsh, Bash, Python REPLs, Git commit messages - -If an application accepts text input and supports accessibility APIs, VocaMac can inject text into it. - -## Beyond Copy-Paste - -Clipboard-based dictation creates unnecessary workflow friction. Your clipboard gets overwritten, breaking your copy-paste rhythm. You need to navigate to a dictation app, paste, then navigate back. This interruption adds up, especially when you're doing multiple short dictations throughout the day. - -With system-wide injection, your workflow stays in your app. Type naturally. Speak naturally. The text appears where you need it, instantly. +VocaMac is designed for browsers, editors, terminals, chat apps, and documents without requiring a separate transcription window. Finish speaking, and the result is inserted at the active cursor through macOS accessibility and keyboard events. -## Configuration Options +## The platform boundary matters -VocaMac respects your preferences for text injection. Configure: +The insertion path works where the foreground app exposes a usable text field and accepts the relevant accessibility events. Secure fields, unusual controls, permission failures, and app-specific input rules can prevent insertion or require a manual paste. VocaMac cannot override those platform decisions. -- **Punctuation handling**: Automatically add periods, commas, or question marks at the end of dictations -- **Capitalization**: Auto-capitalize the first letter of sentences -- **Whitespace**: Add trailing spaces automatically for faster continuation -- **Advanced**: Insert formatting like markdown bold or code blocks +Grant **Accessibility** and **Input Monitoring** access in macOS System Settings. VocaMac's setup and Settings views explain what each permission enables. -These settings apply globally across all applications, ensuring consistent behavior everywhere. +## Clipboard-safe output -## Privacy and Control +VocaMac preserves existing clipboard contents around its paste-based insertion flow. This keeps a copied item available while the transcript is placed into the active field. The result is still local text insertion; it is not a cloud handoff or a per-app integration. -Text injection happens entirely on your device. VocaMac uses only local accessibility APIs provided by macOS. Your transcriptions are never sent to external servers. Your clipboard is never touched. You remain in complete control of what gets typed and when. +## Keep control of the output -Enable accessibility permissions for VocaMac once, then use system-wide dictation in every app on your Mac. No per-app configuration needed. Just speak and create. +VocaMac also lets you configure output details such as capitalization, trailing spaces, and punctuation behavior. Review the inserted text when an app's field or a specialized editor applies its own formatting rules. diff --git a/web/content/features/translation.md b/web/content/features/translation.md index 52ec8d2..0d3887f 100644 --- a/web/content/features/translation.md +++ b/web/content/features/translation.md @@ -1,14 +1,14 @@ --- title: "Translation" -subtitle: "Speak in one language, get text in another. Built-in speech translation powered by Whisper." -description: "VocaMac can translate your speech from one language to another in real-time, all running locally on your Mac. Powered by Whisper's built-in translation capability." +subtitle: "Whisper can translate supported speech into English while it runs locally on your Mac." +description: "VocaMac's Whisper path supports local speech translation into English after the required model is available on your Mac." keywords: "offline speech translation macOS, voice translator mac, speak translate type, whisper translation, local speech translation, voice to text translation mac" icon: "🌐" --- ## Translate While You Speak -Language barriers shouldn't slow down your work. VocaMac's translation feature lets you speak naturally in your native language and have your words transcribed in another language, all processed locally on your Mac with no internet required. +Language barriers shouldn't slow down your work. VocaMac's Whisper path can translate supported speech into English while the model runs locally on your Mac. This is an engine-specific capability, not a general-purpose translation service. This opens possibilities that traditional voice typing simply cannot match: multilingual collaboration, language learning, cross-cultural meetings, and more. All without sending your voice data to cloud servers. @@ -21,15 +21,15 @@ VocaMac uses OpenAI's Whisper model, which has a built-in translation capability 3. **Whisper translates** that transcription to English 4. **VocaMac inserts** the translated English text at your cursor -The entire process happens on your Mac. Your voice never leaves your computer. No cloud calls, no data collection, no privacy concerns. +The speech-processing step happens on your Mac. VocaMac does not send dictation audio to a Voca cloud endpoint; separate actions such as model and release downloads still need a network connection. -Whisper supports translation from over 90 languages into English. While the primary translation target is English, this covers the vast majority of real-world use cases. +The available source languages and results depend on the selected Whisper model. The stable app exposes its supported language hints in Settings; it does not promise every language in Whisper's training data. ## Setting Your Source Language By default, VocaMac automatically detects the language you're speaking. Whisper's detection is remarkably accurate, even for short utterances. -For more predictable results, you can manually set your source language in **Settings → Models → Language**. Simply select from the list of supported languages. Once set, VocaMac will always transcribe in that language (and translate to English if translation is enabled). +For more predictable results, you can manually set your source language in **Settings → Models → Language**. Select from the language hints exposed by the stable app. With a hint selected, VocaMac uses that language for the Whisper path (and translates to English when translation is enabled). Changing languages is as simple as picking a new option from the dropdown. No restarts, no waiting. The change takes effect immediately. @@ -61,20 +61,20 @@ Researchers working with multilingual sources can speak notes in their preferred Support teams, customer success, and engineering teams spanning multiple countries can communicate asynchronously. VocaMac lets each person speak in their native language, with translation happening automatically. Reduces friction, improves clarity, and makes async communication feel more personal. -## Fully Offline Translation +## Local translation after the model download -This is the key advantage of VocaMac's translation feature: everything runs on your Mac. +This is the key boundary of VocaMac's translation feature: once the selected model is available, processing happens locally. Other voice translation services (Google Translate, Microsoft Translator, Amazon Transcribe) require internet connectivity and send your audio or transcription to cloud servers. This introduces latency, battery drain, and privacy concerns. -VocaMac's translation is entirely local. No internet required. No delay. No privacy exposure. The Whisper model sits on your machine and processes everything there. This makes VocaMac suitable for: +VocaMac's translation is local after the model download. The first model download still requires a network connection. This makes VocaMac suitable for: - Offline work environments - Sensitive conversations - Regions with unreliable internet - Users who simply prefer not to transmit their voice data -Once the Whisper model is downloaded (on first run), you can transcribe and translate anywhere, anytime, with complete privacy. +Once the Whisper model is downloaded, the speech-processing step can run without a network connection. Review translated text before using it for high-stakes work. ## Accuracy Considerations @@ -102,7 +102,7 @@ If you want to transcribe in your source language without translation, simply di VocaMac's translation feature represents a fundamentally different approach than cloud-based translation services. By keeping everything local, you get: - **Privacy**: your voice and words never leave your computer -- **Speed**: no network latency, instant processing +- **Speed**: no network round trip for the speech-processing step - **Offline capability**: work anywhere, anytime - **No subscription fees**: translation is included with VocaMac diff --git a/web/content/features/visual-feedback.md b/web/content/features/visual-feedback.md index 234db4e..9a65983 100644 --- a/web/content/features/visual-feedback.md +++ b/web/content/features/visual-feedback.md @@ -1,27 +1,27 @@ --- title: "Visual Feedback" -subtitle: "Menu bar icon changes color. Audio level indicator shows input. Cursor indicator near your caret while recording." -description: "VocaMac provides clear visual feedback during recording with a color-changing menu bar icon, real-time audio level indicator, and an optional floating cursor indicator." +subtitle: "Menu-bar status, audio level, and an optional cursor indicator make recording state visible without relying on color alone." +description: "VocaMac provides clear visual feedback during recording with menu-bar state, status text, an audio level indicator, and an optional floating cursor indicator." keywords: "recording indicator macOS, visual feedback dictation, menu bar recording icon, cursor indicator voice typing, audio level meter mac, recording status indicator" icon: "📊" --- ## Always Know What's Happening -Voice dictation works best when you have complete clarity about your recording state. VocaMac provides three layers of visual feedback so you're never uncertain about whether you're recording, what your audio input looks like, or exactly where your text will appear. +Voice dictation works best when you have complete clarity about your recording state. VocaMac provides several layers of feedback so you can tell whether you are idle, recording, processing, or ready for another dictation. ## Menu Bar Icon ![VocaMac menu bar in idle state](/screenshots/menu-bar-idle.png) ![VocaMac menu bar while recording](/screenshots/menu-bar-recording.png) -The VocaMac menu bar icon is your primary indicator of recording status. It changes color instantly to show what's happening: +The VocaMac menu-bar icon is one indicator of recording status. Its treatment changes instantly, while the popover and status text provide the words that color alone cannot: -- **Blue (idle)**: VocaMac is running and ready to use. No recording is active. This is the default state when you're not speaking. -- **Green (recording)**: You are actively recording. Audio is flowing into the microphone and will be transcribed when you stop. This provides instant visual confirmation that your hotkey press was recognized. -- **Red (error)**: A problem occurred during transcription, audio capture, or model processing. The error message appears in the popover panel with details on how to resolve it. +- **Idle**: VocaMac is running and ready to use. No recording is active. +- **Recording**: Audio is being captured and the popover/indicator shows the active state. +- **Processing or error**: The popover exposes the current status and a useful message when model or audio work needs attention. -These color changes happen instantly. The moment you press your activation hotkey, the icon turns green. The moment you release, it returns to blue. This immediate feedback eliminates the guesswork that plagues other voice apps. +The icon treatment is a quick glanceable cue, while the popover and status text provide the authoritative state. Color is intentionally not the only signal. ## Real-Time Audio Level Indicator @@ -68,7 +68,7 @@ Together, these create a complete feedback loop that matches the directness of k Many users rely on all three feedback sources simultaneously. While recording in a fullscreen text editor: -1. You glance at the menu bar and see the icon is green (recording is active) +1. You glance at the menu bar and see the recording state 2. You see the audio level meter climbing in the popover (your voice is being captured) 3. You see the cursor indicator blinking near your text field (this is where your words will appear) @@ -78,6 +78,6 @@ If any of these signals is missing or unclear, you can adjust settings or check ## Accessibility and Visibility -The color-coded menu bar icon uses high-contrast colors (blue, green, red) that are easily distinguishable. The audio level meter provides both visual and spatial feedback. Together with the cursor indicator, VocaMac ensures that visual feedback is clear and actionable for all users. +The menu-bar state uses icon treatment, status text, and the popover alongside color. The audio level meter provides another signal, while the cursor indicator keeps recording context near the active text field. Together these make feedback clearer without relying on color alone. You control what's visible and when. Customize your feedback experience in Settings to match your preferences and workflow. diff --git a/web/data/product.toml b/web/data/product.toml new file mode 100644 index 0000000..8e7d8a6 --- /dev/null +++ b/web/data/product.toml @@ -0,0 +1,139 @@ +# Centralized VocaMac product facts. +# +# Every figure here was verified against the tagged source of the current +# stable release, the published release artifacts, and the Homebrew cask. +# Re-verify this file before publishing a website change. +# +# Verified: 2026-08-16 against v0.7.2. + +[stable] +version = "0.7.2" +tag = "v0.7.2" +status = "Stable release" +released = "2026-08-04" +releaseURL = "https://github.com/VocaHQ/vocamac/releases/tag/v0.7.2" +releasesURL = "https://github.com/VocaHQ/vocamac/releases" +latestURL = "https://github.com/VocaHQ/vocamac/releases/latest" +dmg = "VocaMac-0.7.2-arm64.dmg" +dmgURL = "https://github.com/VocaHQ/vocamac/releases/download/v0.7.2/VocaMac-0.7.2-arm64.dmg" +dmgSize = "75 MB" + +[requirements] +os = "macOS 13 Ventura or later" +osShort = "macOS 13+" +arch = "Apple Silicon" +archNote = "Apple Silicon only. There is no supported Intel build." +signing = "Developer ID signed and notarized by Apple" +permissions = ["Microphone", "Accessibility", "Input Monitoring"] + +# Languages that can actually be pinned in Settings in v0.7.2, verified by +# counting the tags in the Transcription Language picker in SettingsView.swift. +# The Whisper model is trained on ~99 languages, but only these are selectable, +# so the site must not claim "99+ languages". +[languages] +count = 17 +list = "English, Spanish, French, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean, Hindi, Arabic, Russian, Turkish, Polish, Swedish, and Ukrainian" +autoDetect = true + +[engine] +name = "WhisperKit" +url = "https://github.com/argmaxinc/WhisperKit" +runtime = "CoreML" +modelRepo = "argmaxinc/whisperkit-coreml" +modelRepoURL = "https://huggingface.co/argmaxinc/whisperkit-coreml" + +# The model picker shown in Settings in v0.7.2 (ModelSize.standardCatalog). +# size/ram are the values the app itself reports. +[[models]] +id = "tiny" +name = "Tiny" +size = "39 MB" +ram = "1 GB" +quality = "Good" +note = "Fastest. Good for short commands and quick notes." + +[[models]] +id = "base" +name = "Base" +size = "142 MB" +ram = "1.5 GB" +quality = "Better" +note = "A small step up in accuracy for very little extra weight." + +[[models]] +id = "small" +name = "Small" +size = "466 MB" +ram = "2 GB" +quality = "Great" +note = "A sensible default on 8 GB Macs." + +[[models]] +id = "large-v3-turbo-compact" +name = "Large v3 Turbo (Compact)" +size = "632 MB" +ram = "4 GB" +quality = "Excellent" +note = "The best accuracy-per-second trade-off for most people." + +[[models]] +id = "distil-large-v3-compact" +name = "Distil Large v3 (Compact)" +size = "594 MB" +ram = "4 GB" +quality = "Excellent" +note = "Distilled for speed. English-focused." + +[[models]] +id = "distil-large-v3-turbo" +name = "Distil Large v3 Turbo" +size = "600 MB" +ram = "4 GB" +quality = "Excellent" +note = "The fastest of the distilled variants." + +[[models]] +id = "large-v3-latest-compact" +name = "Large v3 Latest (Compact)" +size = "626 MB" +ram = "5 GB" +quality = "Best" +note = "Full large-model accuracy, compressed." + +[[models]] +id = "large-v3-latest" +name = "Large v3 Latest" +size = "3.1 GB" +ram = "10 GB" +quality = "Best" +note = "Highest accuracy. Wants a 16 GB Mac and patience." + +[links] +repo = "https://github.com/VocaHQ/vocamac" +issues = "https://github.com/VocaHQ/vocamac/issues" +license = "https://github.com/VocaHQ/vocamac/blob/main/LICENSE" +vocahq = "https://vocahq.com/" +homebrewDoc = "https://github.com/VocaHQ/vocamac/blob/main/docs/HOMEBREW.md" + +[license] +id = "AGPL-3.0" +name = "GNU Affero General Public License v3.0" + +[brew] +tap = "brew tap vocahq/vocamac" +trust = "brew trust vocahq/vocamac" +install = "brew install --cask vocamac" +upgrade = "brew upgrade --cask vocamac" + +# Shipped only in nightly builds from main, not in the stable release above. +# Anything listed here must be labelled as nightly wherever it appears. +[nightly] +cask = "brew install --cask vocamac-nightly" +url = "https://github.com/VocaHQ/vocamac/releases/tag/nightly" +features = [ + "Additional speech-to-text engines (Parakeet, Apple Speech, sherpa-onnx)", + "Headless file transcription from the command line", + "A configurable recording overlay", + "Microphone switching from the menu bar", + "Deleting downloaded models from Settings", +] diff --git a/web/hugo.toml b/web/hugo.toml index e4fb925..6fe781c 100644 --- a/web/hugo.toml +++ b/web/hugo.toml @@ -1,14 +1,13 @@ baseURL = "https://vocamac.com/" -languageCode = "en-us" -title = "VocaMac - 100% Local Voice-to-Text for macOS" +locale = "en-us" +title = "VocaMac — native voice typing for macOS" [params] - description = "100% offline, open-source voice-to-text for macOS. Powered by WhisperKit + CoreML." + description = "Native voice typing for macOS. Hold a hotkey, speak, and text appears at your cursor — transcribed by a Whisper model running on your own Mac. Free and open source." author = "Jatin Kumar Malik" twitterHandle = "@jatinkrmalik" githubRepo = "https://github.com/VocaHQ/vocamac" - gaID = "G-8W6EJV2KP0" - ogImage = "https://vocamac.com/og-image.png" + ogImagePath = "og-image.png" [build] [build.buildStats] diff --git a/web/layouts/_default/404.html b/web/layouts/_default/404.html index 66c73a3..ae89cd2 100644 --- a/web/layouts/_default/404.html +++ b/web/layouts/_default/404.html @@ -1,9 +1,10 @@ {{ define "main" }} -
-
-

404

-

Page not found. The page you're looking for doesn't exist or has been moved.

- ← Back to Home -
-
+
+
+

404 / not here

+

That page wandered off.

+

Try the VocaMac home page or browse the feature notes.

+
Back to VocaMacBrowse features
+
+
{{ end }} diff --git a/web/layouts/_default/_markup/render-image.html b/web/layouts/_default/_markup/render-image.html new file mode 100644 index 0000000..729bcb9 --- /dev/null +++ b/web/layouts/_default/_markup/render-image.html @@ -0,0 +1,15 @@ +{{- $dimensions := dict + "/screenshots/cursor-indicator.png" (dict "width" 1217 "height" 990) + "/screenshots/menu-bar-idle.png" (dict "width" 872 "height" 328) + "/screenshots/menu-bar-recording.png" (dict "width" 796 "height" 765) + "/screenshots/popover-panel.png" (dict "width" 852 "height" 508) + "/screenshots/settings-about.png" (dict "width" 1344 "height" 1260) + "/screenshots/settings-audio.png" (dict "width" 1344 "height" 1260) + "/screenshots/settings-general.png" (dict "width" 1344 "height" 1340) + "/screenshots/settings-models.png" (dict "width" 1344 "height" 1260) + "/screenshots/settings-stats.png" (dict "width" 1344 "height" 1340) + "/screenshots/settings-vocabulary.png" (dict "width" 1344 "height" 1340) + "/screenshots/update-ux.png" (dict "width" 1344 "height" 1368) +-}} +{{- $size := index $dimensions .Destination -}} +{{ .Text }} diff --git a/web/layouts/_default/baseof.html b/web/layouts/_default/baseof.html index 4bc24a6..bf43da0 100644 --- a/web/layouts/_default/baseof.html +++ b/web/layouts/_default/baseof.html @@ -5,14 +5,14 @@ {{ block "extra_head" . }}{{ end }} + {{ partial "nav.html" . }} -
+
{{ block "main" . }}{{ end }}
{{ partial "footer.html" . }} - {{ partial "mobile-menu.html" . }} - {{ block "extra_scripts" . }}{{ end }} + diff --git a/web/layouts/_default/list.html b/web/layouts/_default/list.html index c1e822e..50a65a3 100644 --- a/web/layouts/_default/list.html +++ b/web/layouts/_default/list.html @@ -1,25 +1,43 @@ {{ define "main" }} -
-
- ← Back to Home -

{{ .Title }}

- {{ with .Params.description }}

{{ . }}

{{ end }} +{{- $icons := dict + "apple-silicon-native" "spark" + "clipboard-preservation" "clipboard" + "custom-vocabulary" "vocabulary" + "double-tap-toggle" "hotkey" + "fully-configurable" "sliders" + "github-release-updates" "refresh" + "languages" "globe" + "launch-at-login" "launch" + "push-to-talk" "mic" + "silence-detection" "clock" + "smart-model-selection" "chip" + "system-wide-text-injection" "caret" + "translation" "translate" + "usage-stats" "chart" + "visual-feedback" "levels" +-}} +
+
+ Back to VocaMac +

the complete toolbox

+

{{ .Title }}

+ {{ with .Params.description }}

{{ . }}

{{ end }}
-
-
- {{ .Content }} - {{ range .Pages }} -
- -
-
{{ .Params.icon }}
-

{{ .Title }}

-

{{ .Params.subtitle }}

-
+
+
+

VocaMac features

+ {{ with .Content }}
{{ . }}
{{ end }} +
+ {{ range .Pages }} + + {{ partial "icon.html" (default "shield" (index $icons .File.ContentBaseName)) }} +

{{ .Title }}

+

{{ .Params.subtitle }}

+ Read the note + {{ end }}
- {{ end }}
{{ end }} diff --git a/web/layouts/_default/screenshots.html b/web/layouts/_default/screenshots.html index 8d56453..ef1877f 100644 --- a/web/layouts/_default/screenshots.html +++ b/web/layouts/_default/screenshots.html @@ -1,84 +1,54 @@ {{ define "main" }} -
-
- ← Back to Home -

Screenshots

-

See VocaMac in action. From the menu bar to settings to the recording indicator.

-
-
- -
-
- - -
-

📍 Menu Bar

-

VocaMac lives in your menu bar. Always accessible, never in the way.

-
-
- VocaMac menu bar icon in idle state -
Menu bar - Idle
-
-
- VocaMac menu bar icon in recording state -
Menu bar - Recording
-
-
-
- - -
-

🎙️ Popover Panel

-

Click the menu bar icon to see status, last transcription, and quick actions.

-
-
- VocaMac popover panel showing status and last transcription -
Popover panel with status, transcription, and controls
-
-
-
- - -
-

⚙️ Settings

-

Fine-tune everything: hotkeys, models, audio, and more across dedicated tabs.

-
-
- VocaMac Settings - General tab -
General - Activation mode, hotkey, language, and behavior
-
-
- VocaMac Settings - Models tab -
Models - System info, resource usage, and model management
-
- -
- VocaMac Settings - Stats tab -
Stats - Lifetime totals, speaking speed, and daily streaks
-
-
- VocaMac Settings - Audio tab -
Audio - Recording duration, silence detection, and input device
-
-
- VocaMac Settings - About tab -
About - Version info, system details, and links
-
-
-
- - -
-

🔴 Cursor Indicator

-

A floating mic indicator appears near your text cursor while recording, so you always know VocaMac is listening.

-
-
- VocaMac cursor indicator overlay during recording -
Floating mic indicator near the text cursor during recording
-
-
-
+
+
+ Back to VocaMac +

real product states

+

Screenshots

+

See VocaMac in the menu bar, during recording, and inside its settings. These images are product proof, not decorative mockups.

+
+
+
+
+ +
+
+ VocaMac menu bar icon in its idle state +
Menu bar · idleVocaMac stays ready without taking over the desktop.
+
+
+ VocaMac menu bar icon in its recording state +
Menu bar · recordingA visible state change confirms that audio capture is active.
+
+
+ VocaMac popover showing status and the last transcription +
Popover panelCheck readiness, the selected model, and the last result.
+
+
+ VocaMac Settings General tab +
General settingsChoose the activation mode, language, and behavior.
+
+
+ VocaMac Settings Models tab +
Model settingsInspect hardware information and manage local models.
+
+
+ VocaMac Settings Audio tab +
Audio settingsAdjust recording duration, silence detection, and input device.
+
+
+ VocaMac Settings Stats tab +
Local statisticsReview words, speaking speed, and daily activity stored on the Mac.
+
+
+ VocaMac Settings About tab +
About and updatesSee app details and the release update path.
+
+
+ VocaMac cursor indicator beside text during recording +
Cursor indicatorAn optional floating signal keeps the recording state close to the text.
+
-
+
+
{{ end }} diff --git a/web/layouts/_default/single.html b/web/layouts/_default/single.html index 4323b4b..32e6c21 100644 --- a/web/layouts/_default/single.html +++ b/web/layouts/_default/single.html @@ -1,8 +1,15 @@ {{ define "main" }} -
-
-

{{ .Title }}

- {{ .Content }} +
+
+ Back to VocaMac +

VocaMac notes

+

{{ .Title }}

+ {{ with .Params.description }}

{{ . }}

{{ end }} +
+
+
+
+
{{ .Content }}
{{ end }} diff --git a/web/layouts/features/single.html b/web/layouts/features/single.html index 2d22427..78268b5 100644 --- a/web/layouts/features/single.html +++ b/web/layouts/features/single.html @@ -1,24 +1,54 @@ {{ define "main" }} -
-
- ← Back to Home -
{{ .Params.icon }}
-

{{ .Title }}

-

{{ .Params.subtitle }}

+{{- $icons := dict + "apple-silicon-native" "spark" + "clipboard-preservation" "clipboard" + "custom-vocabulary" "vocabulary" + "double-tap-toggle" "hotkey" + "fully-configurable" "sliders" + "github-release-updates" "refresh" + "languages" "globe" + "launch-at-login" "launch" + "push-to-talk" "mic" + "silence-detection" "clock" + "smart-model-selection" "chip" + "system-wide-text-injection" "caret" + "translation" "translate" + "usage-stats" "chart" + "visual-feedback" "levels" +-}} +{{- $icon := default "shield" (index $icons .File.ContentBaseName) -}} +
+
+ All features +
+ {{ partial "icon.html" $icon }} +
+

VocaMac feature

+

{{ .Title }}

+

{{ .Params.subtitle }}

+
-
- -
-
+
+
+
+
+
-
+ + {{ with .TableOfContents }} + + {{ end }} +
+
{{ end }} diff --git a/web/layouts/index.html b/web/layouts/index.html index fb3f20e..e1ffea0 100644 --- a/web/layouts/index.html +++ b/web/layouts/index.html @@ -1,761 +1,315 @@ -{{ define "extra_head" }} - -{{ end }} - {{ define "main" }} - -
-
-
-
- BETA RELEASE - v0.7.2 -
-

- Your Voice,
- Your Mac,
- Your Privacy. -

-

- 100% offline voice-to-text for macOS. Hold a hotkey, speak, and text appears at your cursor. Open source, private by design, and completely free. -

-
-
-
-
-
- - Finder - File - Edit -
-
- VocaMac - Tue 4 Mar -
-
-
-
-
- -
-
VocaMac
-
Model: Small
-
-
-
- 5% - 192 MB -
-
-
-
- ● Ready - Hold Right Option (⌥) -
-
-
-
- Last Transcription - -
-
Hey Siri, just kidding.
VocaMac does it all locally!
-
2.1s audio • 0.2s to transcribe • en
-
-
-
-
-
-
- Quick Install - More Options - Star on GitHub -
- -
- 🔒 100% Offline - 🛡️ Privacy First - 📖 Open Source (AGPL-3.0) - ⚡ Apple Silicon Native -
-
+{{- $p := hugo.Data.product -}} - -
-
-
- Simple -

Three Steps. That's It.

-

No account creation. No cloud configuration. A guided setup wizard gets you started in minutes.

+
+
+
+

native macOS voice typing

+

Say it once.
Your Mac types.

+

VocaMac listens when you ask, transcribes on your Mac, and puts the words at your cursor. No account. Open source. Built for Apple Silicon.

+
+ Get VocaMac + View the source
-
-
-
1
-
⌥
-

Hold Right Option

-

Press and hold your hotkey. The menu bar icon turns red. You're recording.

-
-
-
2
-
🎤
-

Speak

-

Say whatever you want. Your audio is processed entirely on your Mac.

-
-
-
3
-
✨
-

Release

-

Text appears at your cursor. In any app. Browser, Slack, VS Code, anywhere.

-
+
+ {{ $p.requirements.osShort }} + {{ $p.requirements.arch }} + {{ $p.stable.status }} · {{ $p.stable.tag }} + {{ $p.license.id }}
-
- -
-
-
- Features -

Everything You Need,
Nothing You Don't

-

Built for developers and power users who value privacy and speed.

+ -
- - -
-
-
-
- Privacy -

Your Voice Stays
on Your Mac.

-

- Period. No cloud. No servers. No analytics. No telemetry. Not even an internet connection required after installation. -

+
+
+ + + + VocaMac · menu bar
-
-
-
🔒
+
+
-

Zero Network Calls

-

Audio is processed 100% on-device using WhisperKit's CoreML engine. Nothing is ever transmitted.

+ Ready to dictate + Hold Right Option to record
+ local
-
-
🗑️
-
-

No Audio Storage

-

Audio buffers are discarded immediately after transcription. Nothing is saved to disk. Ever.

-
-
-
-
📖
-
-

Fully Open Source

-

Every line of code is public on GitHub. Audit it yourself. AGPL-3.0 licensed. No hidden anything.

-
-
-
-
📡
-
-

No Telemetry

-

No analytics. No crash reporting. No usage tracking. We literally cannot see what you do.

-
-
-
-
-
-

How we compare

-
-
-
-
VocaMac
-
Apple Dictation
-
Others
-
-
-
100% Offline
-
✓
-
~
-
✗
-
-
-
Open Source
-
✓
-
✗
-
✗
-
-
-
No Account Required
-
✓
-
✗
-
✗
+ -
-
Free Forever
-
✓
-
✓
-
✗
+
+ {{/* Space between words only, never after the last one, so the caret sits against the text. */}} + {{ range $i, $word := split "VocaMac puts the words where I am already working" " " }}{{ if $i }} {{ end }}{{ $word }}{{ end }}
-
-
Works in All Apps
-
✓
-
✓
-
~
-
-
-
Custom Models
-
✓
-
✗
-
~
+
+ model: {{ $p.engine.name }} + audio stays on Mac
-
+
+
- -
-
-
- Models -

Choose Your Model

-

From lightning-fast to studio-quality. VocaMac recommends the best one for your Mac.

-
-
-
-
Tiny
-
⚡⚡⚡⚡⚡
-
Good quality
-
~39M params • ~0.4 GB
-
Best for quick notes
-
-
-
Base
-
⚡⚡⚡⚡
-
Better quality
-
~74M params • ~0.8 GB
-
Best for 8 GB Macs
-
-
-
Small
-
⚡⚡⚡
-
Great quality
-
~244M params • ~1.5 GB
-
Best for 16 GB+ Apple Silicon
- Recommended -
-
-
Medium
-
⚡⚡
-
Excellent quality
-
~769M params • ~2.5 GB
-
Best for 24 GB+ high accuracy
+
+
+
+

how it works

+

A small loop that stays out of the way.

+

VocaMac lives in the menu bar, so dictation starts where your work already is.

+
+
+
+ {{ partial "icon.html" "hotkey" }} + 01 / trigger +

Hold a hotkey

+

Use the configured shortcut or toggle mode. A clear menu-bar state tells you when VocaMac is listening.

+

Microphone + Input Monitoring permissions keep the control system-wide.

+
+
+ {{ partial "icon.html" "mic" }} + 02 / speak +

Say what you mean

+

Audio is captured in memory and passed to the speech-to-text model selected in Settings.

+

Choose a language hint when automatic detection needs help.

+
+
+ {{ partial "icon.html" "caret" }} + 03 / insert +

Keep working

+

When recording ends, the transcript is inserted at the cursor through macOS accessibility APIs.

+

Some secure fields and apps can restrict system text insertion.

+
+
+
+
+ +
+
+
+
+

choose your model

+

The model is a setting, not a mystery.

+

The stable release uses {{ $p.engine.name }} with {{ $p.engine.runtime }} acceleration. VocaMac keeps model choice local, visible, and reversible.

+
+

First download

+

Models download from {{ $p.engine.modelRepo }} and are cached locally. After a model is available, transcription can run without sending audio to a cloud service.

-
-
Large v3
-
⚡
-
Best quality
-
~1550M params • ~4.8 GB
-
Best for maximum accuracy
+
+

Nightly is a separate lane

+

The public stable release is {{ $p.stable.tag }}. New engine work may appear in nightly builds before it reaches a stable release; check the release notes before choosing that path.

+
+ + + + + + + {{ range $p.models }} + + + + + + + {{ end }} + +
WhisperKit models available in {{ $p.stable.tag }}
ModelDownloadRAMCharacter
{{ .name }}{{ .size }}{{ .ram }}{{ .note }}
+
-
+
+
- -
-
-
- BETA -

Get Started in Minutes

-

Install via Homebrew, download the DMG, or build from source. Your choice.

-
+
+
+
+

real product states

+

A Mac app you can actually see working.

+

VocaMac is a menu-bar tool, not a dashboard pretending to be one. These are the states that matter while you dictate.

+
+
+
+ VocaMac menu-bar popover showing ready status and the last transcription +
The quiet centerOpen the popover to check status, model, and the last transcript.
+
+
+ VocaMac Settings Models tab showing model management and system information +
Choose your balanceDownload and switch between local models from Settings.
+
+
+ VocaMac recording indicator floating near a text cursor +
Know when it is listeningAn optional cursor indicator keeps recording state visible without taking over your screen.
+
+
+

See the full screenshot gallery

+
+
- -
- - - -
+
+
+
+

a precise privacy promise

+

Your voice has a path.
We show you which one.

+

VocaMac does not need a Voca cloud service to transcribe. The important boundary is the model running on your Mac.

+
- -
-
-
-
1
-
-

Install via Homebrew

-
-
-
- - - -
- Terminal - -
-
- $ brew tap vocahq/vocamac -$ brew trust vocahq/vocamac -$ brew install --cask vocamac -
-
-

Installs to /Applications/VocaMac.app. Launch from Spotlight or your Applications folder.

-
-
-
-
2
-
-

Grant Permissions & Dictate

-

Grant Microphone, Accessibility, and Input Monitoring permissions when prompted. Permissions persist across upgrades — no re-granting needed.

-

Hold Right Option (⌥) → speak → release. Text appears at your cursor. ✨

-
-
+
    +
  1. +
    + {{ partial "icon.html" "mic" }} + Microphone + You hold the hotkey
    -
    -
    -

    Requirements

    -
      -
    • macOS 14 (Sonoma) or later
      • -
    • Apple Silicon (M1/M2/M3/M4)
      • -
    • 8 GB RAM (16 GB+ recommended)
      • -
    • Homebrew installed
      • -
    • Microphone
      • -
    -
    -
    -

    Staying Up to Date

    -

    Run brew upgrade --cask vocamac to get the latest version. VocaMac detects Homebrew installs and guides you automatically. Learn more →

    -
    + +
    2. +
  2. +
    + {{ partial "icon.html" "memory" }} + In-memory audio + Held for the recording
    -
- - - -
+

+ {{ partial "icon.html" "cloud-off" }} + There is no extra stop. The path above has no Voca speech server in it. Dictation audio is not uploaded for transcription. +

- -
-
-
- Ecosystem -

The Voca Family

-

Voice dictation, done right, on every platform.

-
-
- -
- Linux -
-
VocaLinux
-
Beta
-

Voice dictation for Linux. X11 & Wayland. whisper.cpp powered.

- vocalinux.com → - -
-
- Apple -
-
VocaMac
-
Beta
-

Native macOS menu bar app. WhisperKit + CoreML.

- vocamac.com -
-
-
- Windows -
-
VocaWin
-
Coming Soon
-

Voice dictation for Windows. Native system tray app. Planned.

- vocawin.com -
-
+
+
{{ partial "icon.html" "lock" }}

What stays local

During transcription, your audio and the selected model stay on the Mac. VocaMac has no Voca-hosted speech endpoint.

+
{{ partial "icon.html" "globe" }}

What uses a network

Initial model downloads, release downloads, and the app's update check use their documented sources. Apple may manage language assets for system speech.

+
{{ partial "icon.html" "code" }}

What you can inspect

The application is open source under {{ $p.license.id }}. Preferences, local model management, and the injection path are visible in the repository.

-
+
+
- -
-
-
- FAQ -

Questions & Answers

-
-
-
- Is VocaMac really 100% offline? -

Yes. All speech recognition happens locally on your Mac using WhisperKit's CoreML engine. Your voice data never leaves your computer. The only network call is a one-time model download from HuggingFace during setup.

-
-
- Which Macs are supported? -

VocaMac requires an Apple Silicon Mac (M1, M2, M3, or M4) running macOS 14 (Sonoma) or later. The released DMG is built for arm64 only and will not install on Intel Macs. WhisperKit relies on the Apple Neural Engine and CoreML for hardware-accelerated transcription, so Intel Macs are not supported.

-
-
- How does it inject text into other apps? -

VocaMac uses a clipboard-based approach: it saves your clipboard, writes the transcription, simulates Cmd+V to paste, then restores your clipboard. This works in virtually every macOS application.

-
-
- Why does it need Accessibility permission? -

Accessibility permission is required for two things: (1) listening for the global hotkey (Right Option) regardless of which app is focused, and (2) simulating the Cmd+V keystroke to paste text. Without it, hotkeys and text injection won't work.

-
-
- Can I change the hotkey? -

Yes! Open Settings from the menu bar and pick a preset (Right Option, Right Command, Right Control, Fn, or a function key), or click Record and press any key to set a custom one. You can also switch between push-to-talk and double-tap toggle modes.

-
-
- How does this compare to Apple's built-in dictation? -

Apple's dictation sends audio to Apple's servers for processing (for most languages). VocaMac processes everything locally. Your audio never leaves your Mac. You also get model selection, configurable hotkeys, and open-source transparency.

-
-
- Is VocaMac free? -

Yes, completely free and open source under the AGPL-3.0 license. No premium tiers, no subscriptions, no tracking. Just free software.

-
-
- How do I update to the latest version? -

If you installed via Homebrew, run brew upgrade --cask vocamac. VocaMac detects Homebrew installs and guides you automatically. If you installed via DMG, VocaMac checks for updates automatically once every 24 hours, or you can check manually from Settings → About → Check for Updates.

-
-
- What's the difference between WhisperKit and whisper.cpp? -

Both use the same OpenAI Whisper models, so transcription quality is identical. WhisperKit is a pure Swift framework optimized for Apple platforms. It uses CoreML and Neural Engine for hardware acceleration, whereas whisper.cpp uses Metal. VocaLinux uses whisper.cpp; VocaMac uses WhisperKit.

-
-
+
+
+
+

a focused tool, with real controls

+

Make the small details yours.

+

VocaMac keeps its surface compact while giving you control over the parts that change daily use.

+
+
+
{{ partial "icon.html" "hotkey" }}

Hotkeys and modes

Choose push-to-talk or double-tap toggle, then keep a shortcut that fits your hands.

Explore activation
+
{{ partial "icon.html" "chip" }}

Models and language

Pick a local model and let it follow automatic language detection or a language hint.

Explore model choice
+
{{ partial "icon.html" "levels" }}

Microphone and feedback

Choose the input device and use menu-bar, audio-level, or cursor feedback when you need it.

Explore feedback
+
{{ partial "icon.html" "vocabulary" }}

Your vocabulary

Add names and terms that matter to you. Custom vocabulary is processed with the Whisper path.

Explore vocabulary
-
+
+
- -
-
-
-

Ready to Ditch the Cloud?

-

Join the early adopters building the future of private, local voice dictation on macOS.

-
- Get Started - Star on GitHub +
+
+
+

{{ $p.stable.status }} · {{ $p.stable.tag }}

+

Pick your way in.

+

The signed release is ready for {{ $p.requirements.osShort }} and {{ $p.requirements.arch }}. A first model download may need an internet connection.

+
+
+
+ {{ partial "icon.html" "terminal" }} + recommended / Homebrew +

Install from the tap

+

Use the cask when you want upgrades to fit into your normal Mac workflow.

+
+ 
$ brew tap vocahq/vocamac
+$ brew trust vocahq/vocamac
+$ brew install --cask vocamac
-
+

Updates: {{ $p.brew.upgrade }}

+ Read the Homebrew guide + +
+ {{ partial "icon.html" "download" }} + no terminal required / DMG +

Download the signed app

+

Open the latest DMG, drag VocaMac to Applications, and launch it from Spotlight.

+ Download {{ $p.stable.tag }} DMG +

{{ $p.stable.dmg }} · {{ $p.stable.dmgSize }} · {{ $p.requirements.signing }}

+ See releases and checksums +
-
- -{{ end }} - -{{ define "extra_scripts" }} - +
+
+

ready when you are

+

Let your voice stay close to the work.

+

Install VocaMac, choose a model, and put the words where your cursor already is.

+
Get VocaMac See the product
+
{{ $p.requirements.osShort }}{{ $p.requirements.arch }}{{ $p.license.id }}
+
+
{{ end }} diff --git a/web/layouts/partials/footer.html b/web/layouts/partials/footer.html index 3b830db..6b09e5c 100644 --- a/web/layouts/partials/footer.html +++ b/web/layouts/partials/footer.html @@ -1,16 +1,49 @@ - diff --git a/web/layouts/partials/head.html b/web/layouts/partials/head.html index 002aaf6..261cc73 100644 --- a/web/layouts/partials/head.html +++ b/web/layouts/partials/head.html @@ -1,72 +1,76 @@ +{{- $p := hugo.Data.product -}} +{{- $title := cond .IsHome "VocaMac — native voice typing for macOS, transcribed on your Mac" (printf "%s | VocaMac" .Title) -}} +{{- $description := cond (ne .Params.description nil) .Params.description site.Params.description -}} +{{- /* absURL, not a hardcoded domain string: a literal "https://vocamac.com/..." + can't follow a --baseURL override, so a Netlify preview build kept + advertising production's card to link scrapers instead of its own. */ -}} +{{- $ogImage := site.Params.ogImagePath | absURL -}} - - {{ if .IsHome }}VocaMac - 100% Local Voice-to-Text for macOS{{ else }}{{ .Title }} | VocaMac{{ end }} - - - {{ with .Params.keywords }}{{ end }} - {{ if .IsPage }}{{ if not .IsHome }} - - {{ end }}{{ end }} - + {{ $title }} + + {{ with .Params.keywords }}{{ end }} + + - - - - + - - - + + + + + - + - - - - - - + + + + + - - {{ with .Site.Params.gaID }} - - +{{- else if .IsPage }} + - {{ end }} +{{- end }} diff --git a/web/layouts/partials/icon.html b/web/layouts/partials/icon.html new file mode 100644 index 0000000..10c86c9 --- /dev/null +++ b/web/layouts/partials/icon.html @@ -0,0 +1,38 @@ +{{- /* + Inline SVG icon set. + + Usage: {{ partial "icon.html" "mic" }} + + Every glyph is drawn on a 24x24 grid with `currentColor` strokes so it takes + the colour of whatever it sits inside. Icons are decorative: the surrounding + markup always carries the readable label, so each one is aria-hidden. +*/ -}} +{{- $paths := dict + "hotkey" `` + "mic" `` + "caret" `` + "chip" `` + "memory" `` + "levels" `` + "vocabulary" `` + "lock" `` + "globe" `` + "code" `` + "terminal" `` + "download" `` + "cloud-off" `` + "spark" `` + "shield" `` + "translate" `` + "clock" `` + "clipboard" `` + "chart" `` + "launch" `` + "refresh" `` + "sliders" `` + "chevron" `` +-}} +{{- $d := index $paths . -}} +{{- if $d -}} + +{{- end -}} diff --git a/web/layouts/partials/mobile-menu.html b/web/layouts/partials/mobile-menu.html deleted file mode 100644 index 4e4a93e..0000000 --- a/web/layouts/partials/mobile-menu.html +++ /dev/null @@ -1,63 +0,0 @@ - diff --git a/web/layouts/partials/nav.html b/web/layouts/partials/nav.html index 29ca7d2..fc7ea3b 100644 --- a/web/layouts/partials/nav.html +++ b/web/layouts/partials/nav.html @@ -1,24 +1,35 @@ -