Skip to content

CHANGELOG.md

Latest commit

 

History

History
441 lines (405 loc) · 41.5 KB

CHANGELOG.md

File metadata and controls

441 lines (405 loc) · 41.5 KB
Raw

Changelog

Version 0.6.9 – 2026-04-08

Fixes • Improved the Find in Files experience: when opening Browse Folder, focus now goes straight to the folder list; opening a result with Enter no longer breaks keyboard commands; pressing Esc returns to the previously selected result; and when returning with Alt+Tab, focus goes either to the search field or to the results list if results are open. • F5 always started reading from the beginning. It has now been fixed and reading starts from the current caret position, while preserving Shift+F5 and Ctrl+F5 for previous and next sentence navigation. • After using Go to Line, pressing Esc could move focus out of Sonarpad. It now correctly returns focus to the editor. • The Word Wrap option is now applied immediately to documents that are already open, instead of only taking effect after reopening the file.

Version 0.6.8 – 2026-04-07

What's New • Added a new item to the Play menu that lets you transcribe any audio or video file with Whisper. A new “AI and Transcription” section is now available in Options, where you can choose the model, enable optional CUDA support for NVIDIA graphics cards, preserve the original language, and enable or disable timestamps. • Added a new Play menu action, Transcribe current folder, which transcribes all supported audio files in the folder of the currently open media into a single combined document, with dedicated progress, current-file status, and cancellation support. It can also be started with Alt+Shift+C. • Added offline voice dictation, using the same workflow as audio transcription. By default, press Ctrl+Shift+Space to start dictation and press the same shortcut again to stop it; the shortcut can be customized in Options. From the second activation onward, dictation is faster because the engine stays ready in memory; this preloading and reuse are automatically disabled on PCs with less than 4 GB of RAM. • Added a new Editor option, disabled by default, that makes Esc close the editor window. • Podcast search now uses iTunes + Spreaker by default, with duplicate filtering when the same podcast is found on both platforms. • Improved Apple podcast browsing and search: podcast search, category browsing, and top podcasts by category now use the selected podcast directory country. In Options > RSS / Podcast you can leave it on Automatic to use the system country, or choose a different country manually. • Increased the result limit for Apple podcast categories. The first opening still loads the first 50 results as before; if you choose Load more results, Sonarpad loads up to 200 total results (Apple's limit) and lets you navigate through the following pages while keeping the experience smooth. • Sonarpad is now also available on Mac with a subset of features. Project link: https://github.com/Ambro86/Sonarpad-Mac

Improvements • Added more than 50 selectable countries for the podcast directory, so users can choose from a much wider range of national catalogs. • "Play streaming audio..." can now also search YouTube from any text query, or accept a YouTube channel or playlist link and show its results. • Improved how results are shown in "Play streaming audio...": YouTube entries now include title, duration, channel, and view count in a clearer format. • "Play streaming audio..." now also supports YouTube comments: you can open them from the context menu, read replies, and expand comment threads with the Right Arrow key. • Added YouTube favorites for channels and playlists in "Play streaming audio...": they can be added from the results with the context menu, opened directly from the Favorites list reached with Tab right after the YouTube URL/query field, and removed later from that same list using the context menu. In YouTube search results, the context menu is available only for channels and playlists. • "Play streaming audio..." can now request credentials when a streaming site requires login. Users can enter them, save them for the site, and manage saved credentials later in Options > Audio. • Improved focus handling during "Play streaming audio...", so the progress window stays more stable during download and conversion. • Added two new reading navigation actions in the Voice menu: Previous sentence and Next sentence, with configurable shortcuts to jump during text reading. • The default shortcut for Execute file with interpreter is now Ctrl+Shift+F5, so Shift+F5 can be used by default for Previous sentence. • Added voice profiles in Options > Voice: profiles can be added, renamed, and deleted. • Expanded the playback rewind interval options in Options > Audio with additional values from 1 second up to 2 hours. • Added Russian translation thanks to Dmitriy. • Added a new option in Options > Audio to choose the audiobook part naming format: Title + number, Number only, or Number + title. • Added RSS favorite articles: from the article context menu you can add items to a dedicated Favorites feed. • The Favorites RSS feed can be deleted and is recreated automatically when a new article is added to favorites. • Added RSS keyboard shortcuts to move feeds up/down: Ctrl+Shift+Up Arrow and Ctrl+Shift+Down Arrow. • Improved the RSS window with a built-in article preview, so article text can be reviewed directly there and reached quickly with Tab before opening the full article in the editor. • Added an explicit RSS entry “Load more news” at the end of feeds when more items are available; pressing Enter loads the next batch and moves focus to the first newly loaded article. • In the voice dictionary, when adding or editing a replacement, there is now a “Match Case” checkbox so each substitution can either respect or ignore letter casing. Bug fixes • "Play streaming audio..." now respects the podcast cache limit already set in Options, and the same limit now also applies to audio descriptions playback. • Fixed Wikipedia import so quote blocks present in pages are now imported correctly. • Improved the web page parser for WordPress pages where list items and some section headings could be omitted. • "Go to line" now pre-fills the field with the current line. • Fixed OPML export for podcasts and RSS so the exported files are now accepted by iTunes. • Added localized confirmation messages for correct OPML import and export of RSS feeds and podcasts. • Fixed a bug where, in "Play streaming audio...", typing a search string and selecting a YouTube channel from the results could make the program appear stuck instead of opening that channel’s videos. • Fixed a bug where the list of open files was shown in the Help menu instead of the Window menu. • Fixed a streaming edge case where playback could start but the “Downloading stream” dialog stayed open when the downloaded file already matched the target format. • Fixed MP3 streaming conversion behavior: when the stream is already MP3 and the user selects an explicit MP3 bitrate (for example 128 kbps), Sonarpad now re-encodes to the selected bitrate instead of skipping conversion. • Fixed media transcription documents so closing them now asks whether to save, and the suggested file name correctly reuses the transcribed media file name instead of the first line of the text. • Fixed the Alt+Shift+L shortcut: it now correctly opens the chapter list during playback. • Fixed the Alt+Shift+T shortcut: it now correctly starts “Transcribe current audio” instead of opening the Tools menu. • Fixed playback stop handling in the Play menu: pressing . now behaves like Stop and only stops the current track, instead of also exiting the player/episode. • Fixed the save entry in the Play menu for media opened from Recent Files: when the file comes from a local Sonarpad cache, the localized save action is now shown correctly there as well. • When transcription starts while audio is already playing, Sonarpad now pauses that audio automatically before starting transcription. • Fixed a bug where importing an article from Wikipedia could succeed without showing the article text on screen. • Added embedded podcast chapter support from local media files (e.g., MP3 chapter metadata): when feed/URL chapters are unavailable, Sonarpad now loads chapters from the downloaded file in the background, so playback starts immediately and chapter data is applied as soon as it is ready. • Fixed chapter loading for downloaded podcast episodes opened as normal local media files: embedded chapters are now available there too, not only when playback starts from the Podcasts window. • Fixed MP3 audiobook finalization for SAPI4 and SAPI5: final output is now finalized correctly to avoid incomplete or fragile files after long exports. • Added an explicit finalization progress bar for all audiobook creation modes: after the creation phase, Sonarpad now announces and shows a dedicated finalization phase with visible progress. • Fixed dialogue voice tuning: speed/pitch/volume settings are now correctly applied for both the first and second dialogue voices during synthesis. • Improved text encoding detection for Japanese .txt files: added a safe Shift_JIS/CP932 fallback for mojibake cases, while preserving existing UTF/diacritics/Chinese behavior. • Internal safety refactor: converted functions to safe implementations where possible and significantly reduced unsafe code lines.

Version 0.6.7 – 2026-03-02 Improvements • Ora il programma riesce a gestire Sostituisci tutto in modo massivo su file grandi con un gran numero di sostituzioni. • Updated Polish translation thanks to DJ Graco. • Added Lithuanian translation. • Added Chinese translation. • From now on, frequent beta builds will be published in the project Releases section, so users can test new changes before the next stable release. • Added shortcut Ctrl+. to insert an ellipsis character (…). • Improved podcast chapter support: chapter navigation now works more reliably, including direct/streamed episodes where chapters are not embedded in the MP3 file, by using chapter metadata from feed/URL fallbacks when available. Added chapter navigation shortcuts Ctrl+Alt+PageUp (previous chapter) and Ctrl+Alt+PageDown (next chapter). • Reorganized Sonarpad output folders under Documents\\Sonarpad: files are now saved in dedicated subfolders audiobooks, documents, recordings, and media, with automatic migration from legacy paths. • Improved support for very large text files (including 60 MB): smoother opening and line-by-line navigation, especially with screen readers. • Updated guides for all languages and refreshed localization resources across the app, including donations texts and NSIS setup translations (new Simplified Chinese and Lithuanian installer strings, plus completed Ukrainian setup translation). • Added global network proxy support (HTTP/HTTPS and SOCKS5/SOCKS5H) for online features, with proxy validation on Options save: invalid proxies are warned and automatically removed. • Added a new Tools action: "Play streaming audio...", allowing users to paste a URL (YouTube or direct media link), choose output format and quality/bitrate profile (including original quality/bitrate for MP3 and MP4), and play it directly in Sonarpad’s audio player. • Added support for the system media Play/Pause key (headsets/keyboards): it now controls both media playback and text reading pause/resume (with media playback priority when both are active). • Added a new File > Recent Files entry: "Clear recent files" to quickly wipe the recent documents list. • Expanded audio bitrate options in Convert Audio and podcast recording settings: added lower values (64/96 kbps) and extended MP3 up to 320 kbps, with aligned validation and encoder handling. • Extended audiobook split-by-time options up to 60 minutes. • Improved audiobook split-by-parts: users can now enter the number of parts manually, with validation from 1 to 100. • Added a new View > Read-only mode to lock editor text from accidental edits while keeping documents fully readable and navigable. • Added an accessible progress bar during program updates, so screen readers can track download progress in real time. • Added a new quiet status bar in the main window showing characters, words, and line/column (for example: "Characters (with spaces): 11. | Words: 2. | Ln 1, Col 12") without disturbing NVDA focus. • Added a new View menu toggle for Word wrap, so line wrapping can be switched quickly without opening Options. • Added new Edit > Text actions for indent/outdent, with shortcuts Ctrl+Shift+. (indent) and Ctrl+Shift+, (outdent), because when “Show voices in editor” is enabled the Tab key is reserved for voice-panel navigation. • Added localized date/time in RSS articles and podcast episodes, with formatting adapted to the current interface language. • Added a new RSS context-menu action to share the selected article by email. • Added granular delete-confirmation options for RSS/Podcast in Options > RSS and podcast: RSS (feed/article/both/none) and Podcasts (podcast/episode/both/none). • Added configurable quick RSS copy with Ctrl+C (Options > RSS and podcast): copy title, URL, article content, or all combined. • Unified RSS source creation: “Add source” now accepts both direct feed URLs and keyword input (auto-generating Google News RSS), replacing the need for a separate keyword-search action. • Pressing Ctrl+A now announces completion for clearer screen-reader feedback. • Added Shift+F3 for "Find previous" in the Edit menu, complementing F3 "Find next". • Improved replace feedback messages with proper singular/plural forms (e.g. “1 replacement made” vs “2 replacements made”). • Added dictionary lookup language selection in the dictionary window, with default Auto (interface language) and optional manual override. • Added a new Shortcuts tab in Options to customize key bindings, with conflict detection that warns when a shortcut is already assigned to another action. • Added initial command-line switch support: -h/--help now show usage information and --version prints the program version. • Improved manual speed and pitch tuning clarity: manual fields now use a 100-centered scale, where 100 corresponds to the normal value. • Improved Microsoft voices selection in both Options > Voice and the in-editor Voice panel: added a localized language combo to filter voices by language, while keeping multilingual-only mode as a single ungrouped voice list (language combo hidden when enabled). • Added dialogue-voice configuration in Options > Voice with full Tab navigation, using the same voice model as the main UI (engine, Edge language filter, voice, and labeled speed/pitch/volume); added optional secondary dialogue voice with the same controls (engine, Edge language filter, voice, speed/pitch/volume) for alternating dialogues; dialogue voice rules are saved in configuration .ini, so document text is not modified. • Improved Undo labeling: the Edit > Undo entry now shows what action will be undone (for example, text edits, quote/unquote lines, or voice-tag insertion), while remaining disabled when no undo is available. Bug fixes • Fixed RTF file opening: .rtf documents are now parsed and displayed as plain readable text instead of raw RTF markup (e.g. {\\rtf1...}). • Fixed opening of Chinese text files encoded in GB18030/GBK: Sonarpad now detects and decodes these files correctly, avoiding mojibake output. • Improved M4B audiobook creation with chapter metadata and chapter markers; fixed the chipmunk playback issue (high pitch/speed) in generated M4B files. • Fixed audiobook save dialog bitrate UI: removed hardcoded Italian labels and added 64 kbps to selectable bitrate options. • Fixed Save All (Ctrl+Shift+S): all open modified documents are now detected reliably (including unsaved/new tabs), and Save All correctly saves each one or opens Save As where needed. • Fixed Google News RSS item ordering: articles are now shown in descending publish date (newest first) when dates are available. • Fixed NVDA label association in the dictionary window: search field and language combo now announce the correct labels. • Fixed RSS/Podcast Properties window keyboard handling: Tab/Shift+Tab now reaches the OK button, Enter activates OK, Esc closes safely, and focus correctly returns to the RSS/Podcast list. • Fixed RSS/Podcast undo history: Ctrl+Z now supports multi-level undo for removals (articles/episodes and sources), not just the last action. • Improved RSS/Podcast removal feedback with explicit status announcements (RSS removed, RSS article removed, podcast episode removed). • Improved RSS/Podcast focus behavior after delete/undo: RSS now reliably focuses the first feed when needed and avoids repeated screen-reader announcements during delayed reselection.

Version 0.6.6 – 2026-02-13 Improvements • Added "Auto format for TTS" in the Edit menu to quickly prepare text for speech (removes markdown/quotes and reflows wrapped lines). • Improved voice-tag insertion: when text is selected, tags are now applied correctly to both single-line and multi-line selections. • Added a default audiobook save folder option in Audio settings (default: Documents\Sonarpad Audiobooks). • In the audiobook save dialog, when split mode is enabled, added a new default-on option to create a dedicated subfolder for split parts (for cleaner output organization). • Audiobook export now saves MP3 in stereo with user-selected bitrate for Edge, SAPI5, and SAPI4 voices. • Added support for 32-bit SAPI5 voices via bridge, so voices available only in 32-bit engines can also be used in Sonarpad. • Reorganized voice features into a dedicated "Voice and audio" menu and added/clarified "Convert Audio", useful for converting any supported media file to MP3, AAC, OGG, Opus, FLAC, WAV, and AIFF. • Added removal of individual RSS articles and podcast episodes (Delete key + context menu with confirmation), without removing the entire RSS/podcast source, plus undo for the last removal (single article/episode or entire RSS/podcast source). • Added RSS feed export to OPML in the RSS window, so current RSS sources can be saved and re-imported easily. • Added "Search RSS by keyword" in the RSS window: entering a keyword now generates a Google News RSS URL automatically and opens the add-source dialog prefilled, so keyword feeds can be created in one step. • Added Serbian translation thanks to Mila Kuran. • Added Ukrainian translation thanks to Ivan Shtefuriak. • Added multi-file media opening: selecting/opening multiple media files now builds a playback queue instead of replacing the current file. • Added variable seek shortcuts during playback: with a 1-minute base skip, Left/Right seeks 60s, Shift+Left/Right seeks 20s, and Ctrl+Left/Right seeks 3 minutes. • Added previous/next track navigation shortcuts in the player: Ctrl+PageUp and Ctrl+PageDown. • Added "Reset volume" and grouped reset actions into a dedicated "Reset" submenu in Playback, alongside "Reset speed" and "Reset pitch". • Installer improvements: setup.exe now lets users choose between associating all supported file types or selecting extensions manually; MSI now exposes per-extension file-association choices in the feature tree (default remains all enabled). • Added a new "Window" menu with "Open documents..." to quickly switch to any currently open file. • Updated View > Font: replaced the old chooser with a quick submenu of common fonts (Arial, Calibri, Consolas, Segoe UI, Tahoma, Verdana, Times New Roman, Georgia) while preserving the current text size. • Improved RSS/Podcast announcements with a dual status model: source nodes announce "new items" when a feed/podcast has updates, while individual RSS articles and podcast episodes announce "unread"/"unplayed"; this behavior can be disabled from Options. Bug fixes • Fixed EPUB text extraction for books containing inline HTML comments (): chapter text is now parsed correctly instead of being partially or fully skipped. • Fixed Spanish Wiktionary lookups and dictionary cache handling: Spanish entries like "agua" now load correctly, and old "Word not found" cache entries are no longer reused. • Fixed RSS article import character encoding for some Spanish sources (e.g., El Mundo): accented letters and "ñ" are now preserved correctly in the temporary editor. • Fixed ANSI text decoding for Central European files (e.g., Czech/Polish): Sonarpad now better distinguishes UTF-8 vs ANSI and chooses the correct code page (including Windows-1250) to prevent garbled diacritics. • Fixed RSS source persistence for feeds with URL query parameters (e.g., rss.aspx?c=...): these feeds are now saved and restored correctly after restarting Sonarpad. • Fixed opening Google Drive pointer files (.gdoc, .gsheet, .gslides) from Explorer context menu: when direct read fails with “Incorrect function (os error 1)”, Sonarpad now falls back to shell-open so the document still opens correctly. • Fixed legacy Excel 2010 .xls reading: old binary Excel files are now detected and decoded correctly instead of showing garbled text (e.g. ÐÏ_ࡱ...). • Fixed spellcheck announcement flow: misspellings are now announced again when reviewing text later, and the same mistake is reported again if it is deleted and retyped. • Fixed line-based text actions (e.g. Ctrl+Q / Ctrl+Shift+Q, sort/reverse/unique/join lines): selecting a single line with Shift+Down no longer merges or truncates adjacent lines. • Fixed multi-line behavior for line-based text actions (Ctrl+Q / Ctrl+Shift+Q and related line tools): RichEdit selections using CR-only separators are now normalized correctly, so all selected lines are processed without cutting first characters. • Extended TTS input normalization for visible whitespace symbols (␠/U+2420, ␣/U+2423, ␉/U+2409, ␊/U+240A, ␍/U+240D, ␤/U+2424) to prevent repeated paragraph playback with multilingual voices. • Refined Edge TTS text sanitization with a single validation pipeline: weird/invisible spaces are normalized, long punctuation runs (like "...", "!!!", "???") are compacted, and punctuation-only chunks are skipped to prevent playback loops. • Fixed playback time announcement (Ctrl+I) for MP3/podcast streams: current time is now clamped to track duration, and playback is auto-stopped if position runs past the end. • Improved installer localization coverage: setup.exe now includes additional installer languages (Czech, Polish, French, Serbian), while MSI is kept as a single en-US package to avoid release confusion. • Fixed uninstall cleanup for context menu entries: "Open with Sonarpad" is now removed reliably, including legacy registry scenarios. • Fixed SAPI5 pause/resume reliability: F4 pause now works correctly and resume returns to the expected position instead of restarting from the beginning. • Fixed pause + seek + resume flow for media playback: after pausing and seeking with Left/Right, pressing Space now reliably resumes from the current position instead of stopping or restarting from the beginning.

Version 0.6.5 – 2026-02-07 Improvements • Spanish translation improved thanks to Arturo Fernandez Rivas. • Added an option to split EPUB audiobooks by chapters. • RSS imports now use a dedicated temporary tab (localized title); Save As converts it to a normal document. • Screen reader messages are now also sent to JAWS when available. Bug fixes • Reading from the caret (F5) now starts exactly at the cursor. Previously it could start a couple of lines above because the caret offset did not match CRLF/UTF-16 positions. • Fixed a redraw issue where typing over a selection could make earlier text temporarily disappear until the selection moved. • Fixed EPUB chapter parsing so cover or image-only pages no longer produce spoken CSS (e.g., "padding") or "Sconosciuto" titles. • Fixed audiobook time-splitting from EPUBs with Edge TTS failing on empty/oversized chunks ("Edge audio not sent"). • RSS articles now decode HTML entities (e.g., ", &, <, >). • Save/Save As now suggests the existing filename when saving non-overwritable formats (e.g., EPUB) instead of the first line. • Fixed a bug where podcasts with new episodes were not announced as unplayed, and renamed "Unheard" to "Unplayed" for a more professional label.

Version 0.6.4 – 2026-02-05 Improvements • The program has been renamed to Sonarpad to emphasize sound and audio as the key focus. • Added audio track selection in the Playback menu for media files with multiple audio tracks (e.g., MKV files with multiple languages). • Podcasts now clearly indicate unheard episodes with an "Unheard" prefix before the name. • New tag-based voice switching in text. Examples:

  • Microsoft voices (Edge): Hello
  • SAPI5 voices: Hello
  • SAPI4 voices: <voice sapi4 #1>Hello
  • With speed/pitch/volume: Hello • Enriched podcast categories. • Improved PDF reading with automatic fallback to PDFium. • Improved the article parser for cases where content was not read in full. • Added pitch reset to the Playback menu. • Added a context menu option to create an audiobook from the selected text. • Added audiobook splitting by duration, with the ability to choose the first file name. • Localized the author label in article reading (e.g., "by", "di", "par"). • Added indentation options (tabs/spaces with width) and Tab/Shift+Tab indent/outdent on selected lines. • Fixed Markdown cleanup to handle '*' list bullets when bullet preservation is disabled. • Added an option to use the legacy "Novapad" name in the window title and Start Menu shortcuts. Bug fixes • Fixed a bug where SAPI4 audiobooks could be created differently than expected. • Fixed a bug where seeking past the end of a media file restarted playback from the beginning. • Find in Files window: pressing Enter on a result now opens at the correct snippet position, and Esc returns to results. • Options window: improved visual layout across General, Voice, Editor, and Audio tabs to prevent missing or clipped controls. • Fixed a bookmark issue when changing playback speed. • Fixed Podcast Index categories not displaying correctly. • Fixed apostrophes breaking reading by removing separate dialogue reading; voice tags are used instead.

Version 0.6.3 – 2026-01-30 Improvements • Improved microphone detection. • Added instant playback support for all formats. Bug fixes • Fixed crash in podcast categories window.

Version 0.6.2 – 2026-01-30 New features • Added file execution support (Shift+F5). Users can select an interpreter (e.g., python) in Options, search for it on the computer, and pressing Shift+F5 runs the current script. HTML files open in the browser. • Added support for Google Docs pointer files (.gdoc, .gsheet, .gslides), which automatically open in the default browser. • Added support for M4B audiobook format (Apple/AAC). • Added "Show episodes" option in podcast search results context menu to browse and play episodes without subscribing. • Added "Go to Line" feature (Edit menu or Ctrl+J) to quickly jump to a specific line number. • Added context menu options to order RSS feeds and podcasts (alphabetically or by date). • Added Vietnamese default RSS feeds. • Added a microphone test box in the recording dialog to check levels before starting. • Added "Show description" for podcast episodes in the context menu. • Added support for extended audio/video formats via FFmpeg: mkv, avi, mov, m4v, webm, mpg, ts, wmv, flv, vob, 3gp, flac, ogg, wma, aiff. • Added synchronized subtitle reading support (srt, vtt, ass, sub, sbv, lrc, smi) with NVDA or selected voice. The program searches for a subtitle file with the same name as the media file. Added "Import subtitles" and "Remove subtitles" options in the Playback menu for files with different names. • Added file associations for all new supported audio/video formats in the "Open with Sonarpad" context menu. • Added pitch adjustment setting for any file. • Added option in General settings to enable or disable anonymous error reports. Added a menu item in Help to create a diagnostic ZIP file. • Added option to use a different voice for dialogues, both for live reading and audiobook creation. • Added podcast categories browser to explore podcasts by category (business, art, sport, etc.). Improvements • Opening an audio/video file from Explorer now opens the player view directly instead of the text editor. • Removed the OCR prompt for inaccessible PDFs; OCR is now performed automatically to improve speed and user experience. • Improved Accessible Terminal: NVDA reading now remembers the last read line for better continuity. • SAPI 4: Audiobook creation is now fully parallelized and nearly instant. Added a prompt to choose the number of concurrent processes. • SAPI 4: Eliminated the WAV-to-MP3 bottleneck by converting chunks in parallel during synthesis. • SAPI 4: Improved error handling and automatic cleanup of temporary files. • Find dialog: Renamed "Regex" to "Regular expression" for clarity and added missing translations for search options. • M4B Audiobooks: Better output handling; splitting by parts/markers now produces a single M4B file with proper metadata chapters including title and author. • Player: Fixed bookmark and time announcement precision when playback speed is not 1.0x. • Restored Ctrl+Tab and Ctrl+Shift+Tab navigation in Options. • Added an option in the Playback menu to instantly reset speed to Normal (1.0x). • Updated all dependencies to the latest versions for better performance and stability. • Integrated FFmpeg with dynamic DLL loading to ensure compatibility without blocking startup. • Updated podcast download filters to include new audio/video formats. • Prevented Ctrl+S from saving audio/video files to avoid corruption. • Improved YouTube transcript import making it more robust and resilient. • Improved audiobook part splitting robustness, ensuring no text is lost. • Installer is now fully multilingual, supporting Italian, English, Spanish, Portuguese, Swedish, and Vietnamese based on the user's system language. English is the default for unsupported systems. • Podcast categories: pressing Enter on a category now confirms the selection (equivalent to OK button). • Improved hang detection system to avoid false positives when modal dialogs (error messages, "text not found") are open. Fixes • Fixed a bug where the changelog did not open on startup. • Fixed a bug where the OCR prompt did not appear for inaccessible PDFs opened from Explorer. • Fixed a startup bug that could cause loss of focus or window closure immediately after opening. • Fixed a critical bug in regex search that prevented finding text, including issues with "Wrap around" search and "Dot matches newline" option with Windows line endings. Localization • Added Polish translation. • Added French translation. • Added Czech translation (thanks to Radek Žalud and Jiri Holzinger).

Version 0.6.1 – 2026-01-20 Fixes • Fixed a bug where enabling “Show voices in the editor” caused podcast playback to stop. • Fixed an issue where some podcasts could not be added via URL because the URL was being truncated. • Fixed a bug where normal URLs could no longer be added in the RSS feed feature. • Fixed an issue where the Wikipedia language option was shown multiple times across different settings tabs. • Removed the creation of debug files that were incorrectly generated even in release mode. Improvements • Improved support for Microsoft voices, which now use a dedicated playback method with a different user agent. • Added support for MP4 files.

Version 0.6.0 – 2025-01-20 New features • Added spell checker. From the context menu, users can check whether the current word is correct and, if not, get spelling suggestions. • Added podcast import and export via OPML files. • Added Podcast Index search support in addition to iTunes. Users can enter their free API key and secret (generated using only an email address). • Added support for SAPI4 voices, both for real-time reading and audiobook creation. • Added automatic OCR fallback for non-accessible PDFs: when no extractable text is found, the document is recognized via OCR. • Added dictionary support using Wiktionary. Pressing the Applications key shows definitions, and when available, synonyms and translations into other languages. • Added Wikipedia article import with search, result selection, and direct import into the editor. • Added Shift+Enter shortcut in the RSS module to open an article directly in the original website. Improvements • Microphone selection is now always respected by the application. • In the podcast window, pressing Enter on an episode now immediately announces “loading” via NVDA to confirm the action. • In podcast search results, pressing Enter now subscribes to the selected podcast. • Fixed and improved labels for Ctrl+Shift+O and Podcast Ctrl+Shift+P shortcuts. • Playback speed and volume are now saved in settings and persist across all audio files. • Added a dedicated cache folder for podcast episodes. Users can keep episodes via “Keep podcast” in the Playback menu. The cache is automatically cleaned when exceeding the user-defined size (Options → Audio). • Improved RSS article fetching significantly using libcurl impersonation with Chrome and iPhone profiles, ensuring compatibility with ~99% of sites. • Added read / unread state for RSS articles, with clear indication in the RSS list. • Replace All now reports the number of replacements performed. • Added a Delete Podcast button when navigating the podcast library using Tab. Fixes • Removed the redundant “pending update” entry from the Help menu (updates are already handled automatically). • Fixed a bug where pressing Ctrl+S on an opened MP3 file would save and corrupt the file. • Fixed a UI issue where “Batch Audiobooks” was shown as “(B)… Ctrl+Shift+B” (removed redundant label). • Fixed smart quotes: when enabled, normal quotes are now correctly replaced with smart quotes. • Fixed a bug where using “Go to bookmark” reset the playback speed to 1.0. • Fixed an issue where already-downloaded podcast episodes were re-downloaded instead of using the cached version. Keyboard shortcuts • F1 now opens the Help guide. • F2 now checks for updates. • F7 / F8 now jump to the previous or next spelling error. • F9 / F10 now quickly switch between favorite voices. Developer improvements • Errors are no longer silently dropped: all let _ = patterns have been removed, and errors are now explicitly handled (propagated, logged, or handled with fallbacks as appropriate). • The project now fails to compile if there are warnings: both cargo check and cargo clippy must pass cleanly, with lints tightened and allow removed where possible. • Custom implementations such as strlen / wcslen-style helpers have been removed. String and UTF-16 buffer lengths are now derived from Rust-owned data instead of scanning memory. • DLL handling has been cleaned up and consolidated around libloading, avoiding custom loader logic and PE parsing. • Hand-rolled byte parsing helpers were removed; all byte parsing now uses standard from_le_bytes / from_be_bytes on checked slices. These changes reduce unnecessary unsafe usage, eliminate potential undefined behavior, and make the codebase more idiomatic, robust, and maintainable.

Version 0.5.9 - 2025-01-13 New features • Added RSS reordering from the context menu (up/down/to position) with invalid-position checks. • Added an article context menu with open original site and share via WhatsApp, Facebook, and X. • Added Esc shortcut to return from imported articles to the RSS list. • Added podcast mode: search, subscribe, listen; reorder subscriptions; Esc stops playback and returns to the list; Enter on an episode starts playback. • Added playback speed control for podcasts and MP3 files. • Added Ctrl+T to jump to a specific time. • Added a voice preview button after the volume combo. • Added regex find and replace (Notepad++ style). • Added RSS import from OPML and TXT files. • Added an option to enable "Open with Sonarpad" in File Explorer, including portable builds. Improvements • Improved voice speed/pitch/volume selection, respecting TTS max limits. • Various RSS improvements to download all articles without moving NVDA focus during updates. • Improved audio playback with a dedicated menu, Ctrl+I time announce, and volume up to 300%. • Added missing shortcuts for some functions. • Reorganized the Edit menu with a text cleanup submenu. • Reorganized Options into tabs, with Ctrl+Tab and Ctrl+Shift+Tab navigation. • RSS reader now downloads full article content, matching the browser view. Fixes • Fixed Markdown cleanup removing numbers at the start of lines. • Fixed AltGr+Z triggering undo. • Fixed audiobook recording cancellation so it stops quickly. Localization • Added Vietnamese translation (thanks to Anh Đức Nguyễn).

Version 0.5.8 - 2026-01-10 New features • Added volume control for the microphone and system audio when recording podcasts. • Added a new feature to import articles from websites or RSS feeds, including the most important feeds for each language. • Added a function to remove all bookmarks for the current file. • Added a function to remove duplicate lines and duplicate consecutive lines. • Added a function to close all tabs or windows except the current one. • Added a Donations entry in the Help menu for all languages. Improvements • Improved the accessible terminal to prevent some crashes. • Improved and fixed access keys and keyboard shortcuts across the app. • Fixed an issue where closing the audio playback window did not stop playback. • Added confirmation dialogs for important actions (e.g., remove duplicate lines, remove end-of-line hyphens, remove all bookmarks in the current file). No dialog is shown when the action does not apply. • Added the ability to delete RSS feeds/sites from the library by selecting them and pressing Delete. • Added a context menu in the RSS window to edit or delete RSS feeds/sites. • Removed the setting to move settings to the current folder; the app now handles this automatically based on location (if the exe folder is named "sonarpad portable" or the exe is on a removable drive, settings go to the exe folder in config, otherwise %APPDATA%\\Sonarpad, with fallback to the exe config if the preferred folder is not writable).

Version 0.5.7 - 2026-01-05 New features • Added Batch Audiobooks feature to convert multiple files/folders at once. • Added support for Markdown files (.md). • Added file encoding selection when opening text files. • Added option in the accessible terminal to announce new lines with NVDA. Improvements • Audiobook recording now saves natively to MP3 when selected. • User can now choose the position of the "unsaved changes" asterisk (*) in the window title. • Improved the update system robustness across different scenarios. • Added "Remove Hyphens" in Edit menu to fix OCR line-endings.

Version 0.5.6 - 2026-01-04 Fixes Improved Find in Files so pressing Enter opens the file exactly at the selected snippet. Improvements Added PPT/PPTX support (open as text). Opening non-text formats now saves as .txt to avoid formatting corruption (PDF/DOC/DOCX/EPUB/HTML/PPT/PPTX). Added podcast recording from microphone and system audio (File menu, Ctrl+Shift+R).

Version 0.5.5 – 2026-01-03 New features • Added an accessible terminal optimized for large output and screen readers (Ctrl+Shift+P). • Added a setting to save user settings in the current folder (portable mode). Fixes • Improved Find in Files snippets so the preview stays aligned with the match.

Version 0.5.4 – 2026-01-03 Improvements • Fixed Normalize Whitespace (Ctrl+Shift+Enter). • Added HTML/HTM support (open as text).

Version 0.5.3 – 2026-01-02 New features • Added Find in Files. • Added new text tools: Normalize Whitespace, Hard Line Break, and Strip Markdown. • Added Text Statistics (Alt+Y). • Added new list commands in the Edit menu: • Order Items (Alt+Shift+O) • Keep Unique Items (Alt+Shift+K) • Reverse Items (Alt+Shift+Z) • Added Quote / Unquote Lines (Ctrl+Q / Ctrl+Shift+Q). Localization • Added Spanish localization. • Added Portuguese localization. Improvements • When an EPUB file is open, Save now automatically switches to Save As and exports the content as a .txt file to prevent EPUB corruption.

0.5.2 - 2026-01-01

  • Added a changelog.
  • Added open-with-Sonarpad options and file associations for supported files during installation.
  • Improved message localization (errors, dialogs, audiobook export).
  • Added part selection when using "Split audiobook based on text", with a "Require the marker at line start" option.
  • Added YouTube transcript import with language selection, timestamp option, and improved focus handling.

0.5.1 - 2025-12-31

  • Automatic updates with confirmation, improved error handling and notifications.
  • Audiobook export improvements (text-based split, SAPI5/Media Foundation, advanced controls).
  • TTS improvements (pause/resume, replacement dictionary, favorites).
  • View menu and voice/favorites panels, text color and size.
  • Default language from system locale and localization improvements.
  • CI and Windows packaging (artifacts, MSI/NSIS, cache).

0.5.0 - 2025-12-27

  • Modular refactor (editor, file handler, menu, search).
  • Windows build/packaging workflow and README/license updates.
  • Fix TAB navigation in the Help window.

0.5 - 2025-12-27

  • Preliminary version bump.

0.1.0 - 2025-12-25

  • Initial release: project structure and README.