(feat): export JsonMessage (#58)

Author: fern-api[bot]

.fernignore

@@ -12,6 +12,7 @@ src/wrapper
 src/index.ts
 
 # EVI WebSocket
+src/api/resources/empathicVoice/client/Client.ts
 src/api/resources/empathicVoice/resources/chat/index.ts
 src/api/resources/empathicVoice/resources/chat/client
 src/core/websocket

.mock/definition/empathic-voice/__package__.yml

@@ -49,6 +49,11 @@ service:
                     provider: provider
                   ellm_model:
                     allow_short_responses: true
+                  timeouts:
+                    inactivity:
+                      enabled: true
+                    max_duration:
+                      enabled: true
     create-config:
       path: /v0/evi/configs
       method: POST
@@ -75,6 +80,7 @@ service:
               type: optional<list<optional<root.PostedBuiltinTool>>>
               docs: Built-in tool specification for a Config.
             event_messages: optional<root.PostedEventMessageSpecs>
+            timeouts: optional<root.PostedTimeoutSpecs>
       response:
         docs: Created
         type: root.ReturnConfig
@@ -127,6 +133,13 @@ service:
                 on_new_chat:
                   enabled: true
                   text: text
+              timeouts:
+                inactivity:
+                  enabled: true
+                  duration_secs: 1
+                max_duration:
+                  enabled: true
+                  duration_secs: 1
     list-config-versions:
       path: /v0/evi/configs/{id}
       method: GET
@@ -178,6 +191,11 @@ service:
                     provider: provider
                   ellm_model:
                     allow_short_responses: true
+                  timeouts:
+                    inactivity:
+                      enabled: true
+                    max_duration:
+                      enabled: true
     create-config-version:
       path: /v0/evi/configs/{id}
       method: POST
@@ -205,6 +223,7 @@ service:
               type: optional<list<optional<root.PostedBuiltinTool>>>
               docs: Built-in tool specification for a Config.
             event_messages: optional<root.PostedEventMessageSpecs>
+            timeouts: optional<root.PostedTimeoutSpecs>
       response:
         docs: Created
         type: root.ReturnConfig
@@ -258,6 +277,13 @@ service:
                 on_new_chat:
                   enabled: true
                   text: text
+              timeouts:
+                inactivity:
+                  enabled: true
+                  duration_secs: 1
+                max_duration:
+                  enabled: true
+                  duration_secs: 1
     delete-config:
       path: /v0/evi/configs/{id}
       method: DELETE
@@ -354,6 +380,13 @@ service:
                 on_new_chat:
                   enabled: true
                   text: text
+              timeouts:
+                inactivity:
+                  enabled: true
+                  duration_secs: 1
+                max_duration:
+                  enabled: true
+                  duration_secs: 1
     delete-config-version:
       path: /v0/evi/configs/{id}/version/{version}
       method: DELETE
@@ -443,3 +476,10 @@ service:
                 on_new_chat:
                   enabled: true
                   text: text
+              timeouts:
+                inactivity:
+                  enabled: true
+                  duration_secs: 1
+                max_duration:
+                  enabled: true
+                  duration_secs: 1

.mock/definition/empathic-voice/configs.yml

@@ -219,6 +219,7 @@ types:
           unique identifiers will be assigned to face bounding boxes to
           differentiate different faces. If `false`, all faces will be tagged
           with an `unknown` ID.
+        default: false
       min_face_size:
         type: optional<integer>
         docs: >-
@@ -235,6 +236,7 @@ types:
         docs: >-
           Whether to extract and save the detected faces in the artifacts zip
           created by each job.
+        default: false
   FacePrediction:
     properties:
       frame:
@@ -391,13 +393,12 @@ types:
         docs: >-
           If provided, a `POST` request will be made to the URL with the
           generated predictions on completion or the error message on failure.
-        validation:
-          format: url
       notify:
         type: optional<boolean>
         docs: >-
           Whether to send an email notification to the user upon job
           completion/failure.
+        default: false
   InferencePrediction:
     properties:
       file:
@@ -429,13 +430,12 @@ types:
         docs: >-
           If provided, a `POST` request will be made to the URL with the
           generated predictions on completion or the error message on failure.
-        validation:
-          format: url
       notify:
         type: optional<boolean>
         docs: >-
           Whether to send an email notification to the user upon job
           completion/failure.
+        default: false
       files: list<File>
   InferenceResults:
     properties:
@@ -519,6 +519,7 @@ types:
           unique identifiers will be assigned to spoken words to differentiate
           different speakers. If `false`, all speakers will be tagged with an
           `unknown` ID.
+        default: false
   LanguagePrediction:
     properties:
       text:
@@ -585,6 +586,7 @@ types:
           unique identifiers will be assigned to spoken words to differentiate
           different speakers. If `false`, all speakers will be tagged with an
           `unknown` ID.
+        default: false
   NerPrediction:
     properties:
       entity:
@@ -669,6 +671,7 @@ types:
           unique identifiers will be assigned to spoken words to differentiate
           different speakers. If `false`, all speakers will be tagged with an
           `unknown` ID.
+        default: false
   ProsodyPrediction:
     properties:
       text:
@@ -908,13 +911,12 @@ types:
         docs: >-
           If provided, a `POST` request will be made to the URL with the
           generated predictions on completion or the error message on failure.
-        validation:
-          format: url
       notify:
         type: optional<boolean>
         docs: >-
           Whether to send an email notification to the user upon job
           completion/failure.
+        default: false
   CustomModel:
     discriminated: false
     union:
@@ -944,11 +946,10 @@ types:
       task: optional<Task>
       evaluation: optional<EvaluationArgs>
       alternatives: optional<list<Alternative>>
-      callback_url:
-        type: optional<string>
-        validation:
-          format: url
-      notify: optional<boolean>
+      callback_url: optional<string>
+      notify:
+        type: optional<boolean>
+        default: false
   TrainingCustomModel:
     properties:
       id: string
@@ -963,6 +964,7 @@ types:
           unique identifiers will be assigned to spoken words to differentiate
           different speakers. If `false`, all speakers will be tagged with an
           `unknown` ID.
+        default: false
       confidence_threshold:
         type: optional<double>
         docs: >-

.mock/definition/expression-measurement/__package__.yml

@@ -198,6 +198,7 @@ types:
           identifiers will be assigned to face bounding boxes to differentiate
           different faces. If false, all faces will be tagged with an "unknown"
           ID.
+        default: false
       fps_pred:
         type: optional<double>
         docs: >
@@ -327,6 +328,7 @@ types:
 
           Use reset_stream when one audio file is done being processed and you
           do not want context to leak across files.
+        default: false
       raw_text:
         type: optional<boolean>
         docs: >
@@ -336,6 +338,7 @@ types:
           This parameter is useful if you want to send text to be processed by
           the language model, but it cannot be used with other file types like
           audio, image, or video.
+        default: false
       job_details:
         type: optional<boolean>
         docs: >
@@ -348,6 +351,7 @@ types:
 
 
           This parameter is useful to get the unique job ID.
+        default: false
       payload_id:
         type: optional<string>
         docs: >

.mock/definition/expression-measurement/stream.yml

@@ -1,4 +1,4 @@
 {
   "organization" : "hume",
-  "version" : "0.31.9"
+  "version" : "0.31.24"
 }

.mock/fern.config.json

@@ -1,6 +1,6 @@
 {
     "name": "hume",
-    "version": "0.8.1-beta6",
+    "version": "0.8.1-beta7",
     "private": false,
     "repository": "https://github.com/HumeAI/hume-typescript-sdk",
     "main": "./index.js",

package.json

@@ -24,4 +24,5 @@ export interface PostedConfig {
     /** Built-in tool specification for a Config. */
     builtinTools?: (Hume.empathicVoice.PostedBuiltinTool | undefined)[];
     eventMessages?: Hume.empathicVoice.PostedEventMessageSpecs;
+    timeouts?: Hume.empathicVoice.PostedTimeoutSpecs;
 }

src/api/resources/empathicVoice/resources/configs/client/requests/PostedConfig.ts

@@ -20,4 +20,5 @@ export interface PostedConfigVersion {
     /** Built-in tool specification for a Config. */
     builtinTools?: (Hume.empathicVoice.PostedBuiltinTool | undefined)[];
     eventMessages?: Hume.empathicVoice.PostedEventMessageSpecs;
+    timeouts?: Hume.empathicVoice.PostedTimeoutSpecs;
 }

src/api/resources/empathicVoice/resources/configs/client/requests/PostedConfigVersion.ts

@@ -6,7 +6,11 @@
  * When provided, the output is an assistant end message.
  */
 export interface AssistantEnd {
-    /** The type of message sent through the socket; for an Assistant End message, this must be `assistant_end`. */
+    /**
+     * The type of message sent through the socket; for an Assistant End message, this must be `assistant_end`.
+     *
+     * This message indicates the conclusion of the assistant’s response, signaling that the assistant has finished speaking for the current conversational turn.
+     */
     type: "assistant_end";
     /** Used to manage conversational state, correlate frontend and backend data, and persist conversations across EVI sessions. */
     customSessionId?: string;

src/api/resources/empathicVoice/types/AssistantEnd.ts

@@ -6,10 +6,14 @@
  * When provided, the input is spoken by EVI.
  */
 export interface AssistantInput {
-    /** The type of message sent through the socket; for an Assistant Input message, this must be `assistant_input`. */
+    /** The type of message sent through the socket; must be `assistant_input` for our server to correctly identify and process it as an Assistant Input message. */
     type: "assistant_input";
     /** Used to manage conversational state, correlate frontend and backend data, and persist conversations across EVI sessions. */
     customSessionId?: string;
-    /** Text to be synthesized. */
+    /**
+     * Assistant text to synthesize into spoken audio and insert into the conversation.
+     *
+     * EVI uses this text to generate spoken audio using our proprietary expressive text-to-speech model. Our model adds appropriate emotional inflections and tones to the text based on the user’s expressions and the context of the conversation. The synthesized audio is streamed back to the user as an [Assistant Message](/reference/empathic-voice-interface-evi/chat/chat#receive.Assistant%20Message.type).
+     */
     text: string;
 }

src/api/resources/empathicVoice/types/AssistantInput.ts

@@ -8,16 +8,20 @@ import * as Hume from "../../../index";
  * When provided, the output is an assistant message.
  */
 export interface AssistantMessage {
-    /** The type of message sent through the socket; for an Assistant Message, this must be `assistant_message`. */
+    /**
+     * The type of message sent through the socket; for an Assistant Message, this must be `assistant_message`.
+     *
+     * This message contains both a transcript of the assistant’s response and the expression measurement predictions of the assistant’s audio output.
+     */
     type: "assistant_message";
     /** Used to manage conversational state, correlate frontend and backend data, and persist conversations across EVI sessions. */
     customSessionId?: string;
-    /** ID of the assistant message. */
+    /** ID of the assistant message. Allows the Assistant Message to be tracked and referenced. */
     id?: string;
     /** Transcript of the message. */
     message: Hume.empathicVoice.ChatMessage;
     /** Inference model results. */
     models: Hume.empathicVoice.Inference;
-    /** Indicates if this message was constructed from a text input message. */
+    /** Indicates if this message was inserted into the conversation as text from an [Assistant Input message](/reference/empathic-voice-interface-evi/chat/chat#send.Assistant%20Input.text). */
     fromText: boolean;
 }

src/api/resources/empathicVoice/types/AssistantMessage.ts

@@ -5,10 +5,10 @@
 import * as Hume from "../../../index";
 
 export interface AudioConfiguration {
-    /** Audio encoding. */
+    /** Encoding format of the audio input, such as `linear16`. */
     encoding: Hume.empathicVoice.Encoding;
-    /** Number of channels. */
+    /** Number of audio channels. */
     channels: number;
-    /** Audio sample rate. */
+    /** Audio sample rate. Number of samples per second in the audio input, measured in Hertz. */
     sampleRate: number;
 }

src/api/resources/empathicVoice/types/AudioConfiguration.ts

@@ -6,10 +6,20 @@
  * When provided, the input is audio.
  */
 export interface AudioInput {
-    /** The type of message sent through the socket; for an Audio Input message, this must be `audio_input`. */
+    /**
+     * The type of message sent through the socket; must be `audio_input` for our server to correctly identify and process it as an Audio Input message.
+     *
+     * This message is used for sending audio input data to EVI for processing and expression measurement. Audio data should be sent as a continuous stream, encoded in Base64.
+     */
     type: "audio_input";
     /** Used to manage conversational state, correlate frontend and backend data, and persist conversations across EVI sessions. */
     customSessionId?: string;
-    /** Base64 encoded audio input. */
+    /**
+     * Base64 encoded audio input to insert into the conversation.
+     *
+     * The audio input must be captured and transmitted to EVI as a continuous stream, with the audio data sent in small chunks for better transcription quality. When capturing audio through the browser, we recommend recording the audio in 100ms intervals and adjusting from there to determine if smaller or larger chunks are needed. These chunks should be continuously sent to EVI as Audio Input messages.
+     *
+     * The content of an Audio Input message is treated as the user’s speech to EVI. EVI processes the audio, conducts expression measurement using the prosody model, and responds accordingly.
+     */
     data: string;
 }

src/api/resources/empathicVoice/types/AudioInput.ts

@@ -10,8 +10,8 @@ export interface AudioOutput {
     type: "audio_output";
     /** Used to manage conversational state, correlate frontend and backend data, and persist conversations across EVI sessions. */
     customSessionId?: string;
-    /** ID of the audio output. */
+    /** ID of the audio output. Allows the Audio Output message to be tracked and referenced. */
     id: string;
-    /** Base64 encoded audio output. */
+    /** Base64 encoded audio output. This encoded audio is transmitted to the client, where it can be decoded and played back as part of the user interaction. */
     data: string;
 }

src/api/resources/empathicVoice/types/AudioOutput.ts

@@ -2,4 +2,7 @@
  * This file was auto-generated by Fern from our API Definition.
  */
 
+/**
+ * Name of the built-in tool. Set to `web_search` to equip EVI with the built-in Web Search tool.
+ */
 export type BuiltInTool = "web_search";

src/api/resources/empathicVoice/types/BuiltInTool.ts

@@ -6,5 +6,6 @@ import * as Hume from "../../../index";
 
 export interface BuiltinToolConfig {
     name: Hume.empathicVoice.BuiltInTool;
+    /** Optional text passed to the supplemental LLM if the tool call fails. The LLM then uses this text to generate a response back to the user, ensuring continuity in the conversation. */
     fallbackContent?: string;
 }