> ## Documentation Index > Fetch the complete documentation index at: https://docs.gladia.io/llms.txt > Use this file to discover all available pages before exploring further. # Live WebSocket > WebSocket to send audio chunks and receive events. ## AsyncAPI ````yaml asyncapi.yaml liveTranscription id: liveTranscription title: Live transcription description: Single WebSocket channel for all client->server and server->client messages. servers: - id: real-time protocol: wss host: api.gladia.io/v2/live?token={SESSION} bindings: [] variables: [] address: liveTranscription parameters: [] bindings: [] operations: - &ref_10 id: sendAudioChunkJson title: Send audio chunk json description: Send an audio chunk as JSON (base64). type: receive messages: - &ref_29 id: audioChunkJson contentType: application/json payload: - name: Audio Chunk (JSON) description: Send audio bytes as base64-encoded chunk. type: object properties: - name: type type: string description: audio_chunk required: true - name: data type: object required: true properties: - name: chunk type: string description: Base64-encoded raw audio bytes. required: false headers: [] jsonPayloadSchema: type: object required: - type - data properties: type: type: string const: audio_chunk x-parser-schema-id: data: type: object required: - chunk properties: chunk: type: string description: Base64-encoded raw audio bytes. x-parser-schema-id: x-parser-schema-id: x-parser-schema-id: title: Audio Chunk (JSON) description: Send audio bytes as base64-encoded chunk. example: |- { "type": "audio_chunk", "data": { "chunk": "UklGRiQAAABXQVZFZm10IBIAAAABAAEAESsAACJWAAACABAAZGF0YQAAAA==" } } bindings: [] extensions: - id: x-parser-unique-object-id value: audioChunkJson bindings: [] extensions: &ref_0 - id: x-parser-unique-object-id value: liveTranscription - &ref_11 id: sendAudioChunkBinary title: Send audio chunk binary description: Send an audio chunk as a binary frame. type: receive messages: - &ref_30 id: audioChunkBinary contentType: application/octet-stream payload: - type: string format: binary description: Send audio bytes as a binary WebSocket frame. x-parser-schema-id: name: Audio Chunk (Binary frame) headers: [] jsonPayloadSchema: type: string format: binary description: >- Raw audio bytes matching the session encoding/bit_depth/sample_rate/channels. x-parser-schema-id: title: Audio Chunk (Binary frame) description: Send audio bytes as a binary WebSocket frame. example: '{}' bindings: [] extensions: - id: x-parser-unique-object-id value: audioChunkBinary bindings: [] extensions: *ref_0 - &ref_12 id: sendStopRecording title: Send stop recording description: Stop recording. type: receive messages: - &ref_31 id: stopRecording contentType: application/json payload: - name: Stop Recording description: Inform Gladia that recording is over. type: object properties: - name: type type: string description: stop_recording required: true headers: [] jsonPayloadSchema: type: object required: - type properties: type: type: string const: stop_recording x-parser-schema-id: x-parser-schema-id: title: Stop Recording description: Inform Gladia that recording is over. example: |- { "type": "stop_recording" } bindings: [] extensions: - id: x-parser-unique-object-id value: stopRecording bindings: [] extensions: *ref_0 - &ref_13 id: onTranscript title: On transcript type: send messages: - &ref_32 id: transcript contentType: application/json payload: - name: Transcript description: >- This message contains the informations about the transcript of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: type type: string description: transcript required: true - name: data type: object required: true properties: - name: id type: string required: false - name: is_final type: boolean description: Indicates if the transcript is final or partial. required: false - name: utterance type: object description: >- Continuous segment of speech with timing, text and optional channel. required: false properties: - name: start type: number description: >- Start time of the utterance in seconds from session start. required: false - name: end type: number description: >- End time of the utterance in seconds from session start. required: false - name: confidence type: number description: Confidence score for the utterance between 0 and 1. required: false - name: channel type: integer description: >- Audio channel index (0-based) if multiple channels are present. required: false - name: words type: array description: Word-level details for this utterance. required: false - name: text type: string description: The transcribed text for the utterance. required: false - name: language type: string description: >- BCP-47 language tag detected for the utterance (e.g. "en"). required: false headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - data properties: session_id: &ref_1 type: string format: uuid description: UUID identifying the session. examples: - 550e8400-e29b-41d4-a716-446655440000 x-parser-schema-id: UUID created_at: &ref_2 type: string format: date-time description: ISO 8601 timestamp in UTC. examples: - '2025-09-19T12:34:56Z' x-parser-schema-id: ISODateTime type: type: string const: transcript x-parser-schema-id: data: type: object required: - id - is_final - utterance properties: id: type: string x-parser-schema-id: is_final: type: boolean description: Indicates if the transcript is final or partial. x-parser-schema-id: utterance: &ref_4 type: object properties: start: type: number description: >- Start time of the utterance in seconds from session start. x-parser-schema-id: end: type: number description: End time of the utterance in seconds from session start. x-parser-schema-id: confidence: type: number description: Confidence score for the utterance between 0 and 1. x-parser-schema-id: channel: type: integer description: >- Audio channel index (0-based) if multiple channels are present. x-parser-schema-id: words: type: array items: &ref_9 type: object properties: word: type: string description: Recognized token as spoken in the audio. x-parser-schema-id: start: type: number description: >- Start time of the word in seconds from session start. x-parser-schema-id: end: type: number description: >- End time of the word in seconds from session start. x-parser-schema-id: confidence: type: number description: Confidence score for the token between 0 and 1. x-parser-schema-id: description: >- Word-level timing and confidence for a token in the transcript. examples: - word: hello start: 0.12 end: 0.38 confidence: 0.97 x-parser-schema-id: Word description: Word-level details for this utterance. x-parser-schema-id: text: type: string description: The transcribed text for the utterance. x-parser-schema-id: language: type: string description: >- BCP-47 language tag detected for the utterance (e.g. "en"). x-parser-schema-id: description: >- Continuous segment of speech with timing, text and optional channel. examples: - start: 1.02 end: 3.44 confidence: 0.93 channel: 0 words: - word: good start: 1.02 end: 1.21 confidence: 0.96 - word: morning start: 1.22 end: 1.55 confidence: 0.95 text: good morning language: en x-parser-schema-id: Utterance x-parser-schema-id: description: >- Realtime transcript update for an utterance; may be interim or final. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:34:10Z' type: transcript data: id: 00-00000011 is_final: true utterance: start: 0 end: 0.48 confidence: 0.91 channel: 0 words: - word: Hello start: 0 end: 0.35 confidence: 0.91 - word: world start: 0.36 end: 0.48 confidence: 0.91 text: Hello world. language: en x-parser-schema-id: TranscriptMessage title: Transcript description: >- This message contains the informations about the transcript of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:34:10Z", "type": "transcript", "data": { "id": "00-00000011", "is_final": true, "utterance": { "start": 0, "end": 0.48, "confidence": 0.91, "channel": 0, "words": [ { "word": "Hello", "start": 0, "end": 0.35, "confidence": 0.91 }, { "word": "world", "start": 0.36, "end": 0.48, "confidence": 0.91 } ], "text": "Hello world.", "language": "en" } } } bindings: [] extensions: - id: x-parser-unique-object-id value: transcript bindings: [] extensions: *ref_0 - &ref_14 id: onSpeechStart title: On speech start type: send messages: - &ref_33 id: speechStart contentType: application/json payload: - name: Speech Start description: >- Indicates the start of detected speech activity on a given channel. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: type type: string description: speech_start required: true - name: data type: object description: Timing information for speech start/end events. required: true properties: - name: time type: number description: >- Time in seconds from session start when the event occurred. required: false - name: channel type: integer description: Audio channel index (0-based) for which the event applies. required: false headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - data properties: session_id: *ref_1 created_at: *ref_2 type: type: string const: speech_start x-parser-schema-id: data: &ref_3 type: object properties: time: type: number description: Time in seconds from session start when the event occurred. x-parser-schema-id: channel: type: integer description: Audio channel index (0-based) for which the event applies. x-parser-schema-id: description: Timing information for speech start/end events. examples: - time: 1.24 channel: 0 x-parser-schema-id: SpeechEventData description: Indicates the start of detected speech activity on a given channel. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:34:06Z' type: speech_start data: time: 1.24 channel: 0 x-parser-schema-id: SpeechEventMessage title: Speech Start description: Indicates the start of detected speech activity on a given channel. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:34:06Z", "type": "speech_start", "data": { "time": 1.24, "channel": 0 } } bindings: [] extensions: - id: x-parser-unique-object-id value: speechStart bindings: [] extensions: *ref_0 - &ref_15 id: onSpeechEnd title: On speech end type: send messages: - &ref_34 id: speechEnd contentType: application/json payload: - name: Speech End description: Indicates the end of detected speech activity. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: type type: string description: speech_end required: true - name: data type: object description: Timing information for speech start/end events. required: true properties: - name: time type: number description: >- Time in seconds from session start when the event occurred. required: false - name: channel type: integer description: Audio channel index (0-based) for which the event applies. required: false headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - data properties: session_id: *ref_1 created_at: *ref_2 type: type: string const: speech_end x-parser-schema-id: data: *ref_3 description: Indicates the end of detected speech activity. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:34:09Z' type: speech_end data: time: 3.1 channel: 0 x-parser-schema-id: SpeechEventMessageEnd title: Speech End description: Indicates the end of detected speech activity. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:34:09Z", "type": "speech_end", "data": { "time": 3.1, "channel": 0 } } bindings: [] extensions: - id: x-parser-unique-object-id value: speechEnd bindings: [] extensions: *ref_0 - &ref_16 id: onTranslation title: On translation type: send messages: - &ref_35 id: translation contentType: application/json payload: - name: Translation description: >- This message contains the informations about the translation of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: 'null' required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: integer required: false - name: type type: string description: string required: false - name: description type: string description: HTTP-like code or provider-specific status string. required: false - name: exception type: string description: Short machine-readable error name. required: false - name: message type: string description: Human-readable error explanation. required: false - name: type type: string description: translation required: true - name: data type: object required: true properties: - name: utterance_id type: string required: false - name: utterance type: object description: >- Continuous segment of speech with timing, text and optional channel. required: false properties: - name: start type: number description: >- Start time of the utterance in seconds from session start. required: false - name: end type: number description: >- End time of the utterance in seconds from session start. required: false - name: confidence type: number description: Confidence score for the utterance between 0 and 1. required: false - name: channel type: integer description: >- Audio channel index (0-based) if multiple channels are present. required: false - name: words type: array description: Word-level details for this utterance. required: false - name: text type: string description: The transcribed text for the utterance. required: false - name: language type: string description: >- BCP-47 language tag detected for the utterance (e.g. "en"). required: false - name: original_language type: string required: false - name: target_language type: string required: false - name: translated_utterance type: object description: >- Continuous segment of speech with timing, text and optional channel. required: false properties: - name: start type: number description: >- Start time of the utterance in seconds from session start. required: false - name: end type: number description: >- End time of the utterance in seconds from session start. required: false - name: confidence type: number description: Confidence score for the utterance between 0 and 1. required: false - name: channel type: integer description: >- Audio channel index (0-based) if multiple channels are present. required: false - name: words type: array description: Word-level details for this utterance. required: false - name: text type: string description: The transcribed text for the utterance. required: false - name: language type: string description: >- BCP-47 language tag detected for the utterance (e.g. "en"). required: false headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - data properties: session_id: *ref_1 created_at: *ref_2 error: oneOf: - type: 'null' x-parser-schema-id: - &ref_5 type: object properties: status_code: oneOf: - type: integer x-parser-schema-id: - type: string x-parser-schema-id: description: HTTP-like code or provider-specific status string. x-parser-schema-id: exception: type: string description: Short machine-readable error name. x-parser-schema-id: message: type: string description: Human-readable error explanation. x-parser-schema-id: description: >- Standard error container included in payloads when an operation fails. examples: - status_code: 400 exception: InvalidArgument message: Unsupported sample rate x-parser-schema-id: ErrorObject x-parser-schema-id: type: type: string const: translation x-parser-schema-id: data: type: object required: - utterance_id - utterance - original_language - target_language - translated_utterance properties: utterance_id: type: string x-parser-schema-id: utterance: *ref_4 original_language: type: string x-parser-schema-id: target_language: type: string x-parser-schema-id: translated_utterance: *ref_4 x-parser-schema-id: description: >- Translation result for an utterance into the requested target language. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:34:20Z' type: translation error: null data: utterance_id: utt_001 utterance: text: buenos días language: es start: 4.2 end: 6.1 original_language: es target_language: en translated_utterance: text: good morning language: en start: 4.2 end: 6.1 x-parser-schema-id: TranslationMessage title: Translation description: >- This message contains the informations about the translation of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:34:20Z", "type": "translation", "error": null, "data": { "utterance_id": "utt_001", "utterance": { "text": "buenos días", "language": "es", "start": 4.2, "end": 6.1 }, "original_language": "es", "target_language": "en", "translated_utterance": { "text": "good morning", "language": "en", "start": 4.2, "end": 6.1 } } } bindings: [] extensions: - id: x-parser-unique-object-id value: translation bindings: [] extensions: *ref_0 - &ref_17 id: onNamedEntityRecognition title: On named entity recognition type: send messages: - &ref_36 id: namedEntityRecognition contentType: application/json payload: - name: Named Entity Recognition description: >- This message contains the informations about the named entity recognition of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: 'null' required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: integer required: false - name: type type: string description: string required: false - name: description type: string description: HTTP-like code or provider-specific status string. required: false - name: exception type: string description: Short machine-readable error name. required: false - name: message type: string description: Human-readable error explanation. required: false - name: type type: string description: named_entity_recognition required: true - name: data type: object required: true properties: - name: utterance_id type: string required: false - name: utterance type: object description: >- Continuous segment of speech with timing, text and optional channel. required: false properties: - name: start type: number description: >- Start time of the utterance in seconds from session start. required: false - name: end type: number description: >- End time of the utterance in seconds from session start. required: false - name: confidence type: number description: Confidence score for the utterance between 0 and 1. required: false - name: channel type: integer description: >- Audio channel index (0-based) if multiple channels are present. required: false - name: words type: array description: Word-level details for this utterance. required: false - name: text type: string description: The transcribed text for the utterance. required: false - name: language type: string description: >- BCP-47 language tag detected for the utterance (e.g. "en"). required: false - name: results type: array required: false headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - data properties: session_id: *ref_1 created_at: *ref_2 error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: type: type: string const: named_entity_recognition x-parser-schema-id: data: type: object required: - utterance_id - utterance - results properties: utterance_id: type: string x-parser-schema-id: utterance: *ref_4 results: type: array items: type: object properties: entity_type: type: string x-parser-schema-id: text: type: string x-parser-schema-id: start: type: number x-parser-schema-id: end: type: number x-parser-schema-id: x-parser-schema-id: x-parser-schema-id: x-parser-schema-id: description: Named entity recognition results extracted from the given utterance. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:34:30Z' type: named_entity_recognition error: null data: utterance_id: utt_002 utterance: text: meeting with Alice at 3pm start: 7 end: 10.5 results: - entity_type: PERSON text: Alice start: 2 end: 2.5 - entity_type: TIME text: 3pm start: 3 end: 3.3 x-parser-schema-id: NamedEntityRecognitionMessage title: Named Entity Recognition description: >- This message contains the informations about the named entity recognition of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:34:30Z", "type": "named_entity_recognition", "error": null, "data": { "utterance_id": "utt_002", "utterance": { "text": "meeting with Alice at 3pm", "start": 7, "end": 10.5 }, "results": [ { "entity_type": "PERSON", "text": "Alice", "start": 2, "end": 2.5 }, { "entity_type": "TIME", "text": "3pm", "start": 3, "end": 3.3 } ] } } bindings: [] extensions: - id: x-parser-unique-object-id value: namedEntityRecognition bindings: [] extensions: *ref_0 - &ref_18 id: onSentimentAnalysis title: On sentiment analysis type: send messages: - &ref_37 id: sentimentAnalysis contentType: application/json payload: - name: Sentiment Analysis description: >- This message contains the informations about the sentiment analysis of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: 'null' required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: integer required: false - name: type type: string description: string required: false - name: description type: string description: HTTP-like code or provider-specific status string. required: false - name: exception type: string description: Short machine-readable error name. required: false - name: message type: string description: Human-readable error explanation. required: false - name: type type: string description: sentiment_analysis required: true - name: data type: object required: true properties: - name: utterance_id type: string required: false - name: utterance type: object description: >- Continuous segment of speech with timing, text and optional channel. required: false properties: - name: start type: number description: >- Start time of the utterance in seconds from session start. required: false - name: end type: number description: >- End time of the utterance in seconds from session start. required: false - name: confidence type: number description: Confidence score for the utterance between 0 and 1. required: false - name: channel type: integer description: >- Audio channel index (0-based) if multiple channels are present. required: false - name: words type: array description: Word-level details for this utterance. required: false - name: text type: string description: The transcribed text for the utterance. required: false - name: language type: string description: >- BCP-47 language tag detected for the utterance (e.g. "en"). required: false - name: results type: array required: false headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - data properties: session_id: *ref_1 created_at: *ref_2 error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: type: type: string const: sentiment_analysis x-parser-schema-id: data: type: object required: - utterance_id - utterance - results properties: utterance_id: type: string x-parser-schema-id: utterance: *ref_4 results: type: array items: type: object properties: sentiment: type: string x-parser-schema-id: emotion: type: string x-parser-schema-id: text: type: string x-parser-schema-id: start: type: number x-parser-schema-id: end: type: number x-parser-schema-id: channel: type: integer x-parser-schema-id: x-parser-schema-id: x-parser-schema-id: x-parser-schema-id: description: >- Sentiment and emotion analysis associated with segments of an utterance. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:34:35Z' type: sentiment_analysis error: null data: utterance_id: utt_003 utterance: text: I love this product! start: 11 end: 13.2 results: - sentiment: positive emotion: joy text: love this product start: 11.4 end: 12.9 channel: 0 x-parser-schema-id: SentimentAnalysisMessage title: Sentiment Analysis description: >- This message contains the informations about the sentiment analysis of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:34:35Z", "type": "sentiment_analysis", "error": null, "data": { "utterance_id": "utt_003", "utterance": { "text": "I love this product!", "start": 11, "end": 13.2 }, "results": [ { "sentiment": "positive", "emotion": "joy", "text": "love this product", "start": 11.4, "end": 12.9, "channel": 0 } ] } } bindings: [] extensions: - id: x-parser-unique-object-id value: sentimentAnalysis bindings: [] extensions: *ref_0 - &ref_19 id: onPostTranscript title: On post transcript type: send messages: - &ref_38 id: postTranscript contentType: application/json payload: - name: Post Transcript description: >- This message contains the informations about the post transcript of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: type type: string description: post_transcript required: true - name: data type: object description: >- Aggregated transcription artifacts produced after the session completes. required: true properties: - name: full_transcript type: string required: false - name: languages type: array required: false - name: sentences type: array required: false - name: subtitles type: array required: false - name: utterances type: array required: false headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - data properties: session_id: *ref_1 created_at: *ref_2 type: type: string const: post_transcript x-parser-schema-id: data: &ref_6 type: object properties: full_transcript: type: string x-parser-schema-id: languages: type: array items: type: string x-parser-schema-id: x-parser-schema-id: sentences: type: array items: &ref_7 type: object properties: success: type: boolean x-parser-schema-id: is_empty: type: boolean x-parser-schema-id: exec_time: type: number x-parser-schema-id: error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: results: type: array items: type: string x-parser-schema-id: x-parser-schema-id: description: >- Result container for sentence-level operations such as punctuation or translation. examples: - success: true is_empty: false exec_time: 0.12 error: null results: - Hello, world. x-parser-schema-id: SentenceResult x-parser-schema-id: subtitles: type: array items: &ref_8 type: object properties: format: type: string enum: - srt - vtt - ass x-parser-schema-id: subtitles: type: string x-parser-schema-id: description: Subtitle payload for a specific format. examples: - format: vtt subtitles: | WEBVTT 00:00:00.000 --> 00:00:01.000 Hello world x-parser-schema-id: SubtitleBlock x-parser-schema-id: utterances: type: array items: *ref_4 x-parser-schema-id: description: >- Aggregated transcription artifacts produced after the session completes. examples: - full_transcript: hello world languages: - en sentences: - success: true is_empty: false exec_time: 0.11 error: null results: - Hello world. subtitles: - format: srt subtitles: | 1 00:00:00,000 --> 00:00:01,000 Hello world utterances: - start: 0 end: 0.9 text: hello world x-parser-schema-id: PostTranscriptData description: >- Post-processing transcript payload containing full text and artifacts. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:40:00Z' type: post_transcript data: full_transcript: hello world x-parser-schema-id: PostTranscriptMessage title: Post Transcript description: >- This message contains the informations about the post transcript of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:40:00Z", "type": "post_transcript", "data": { "full_transcript": "hello world" } } bindings: [] extensions: - id: x-parser-unique-object-id value: postTranscript bindings: [] extensions: *ref_0 - &ref_20 id: onPostFinalTranscript title: On post final transcript type: send messages: - &ref_39 id: postFinalTranscript contentType: application/json payload: - name: Final transcript (aggregated) description: >- This message contains the informations about the post final transcript of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: type type: string description: post_final_transcript required: true - name: data type: object required: true properties: - name: metadata type: object required: false properties: - name: audio_duration type: number required: false - name: number_of_distinct_channels type: integer required: false - name: billing_time type: number required: false - name: transcription_time type: number required: false - name: transcription type: object description: >- Aggregated transcription artifacts produced after the session completes. required: false properties: - name: full_transcript type: string required: false - name: languages type: array required: false - name: sentences type: array required: false - name: subtitles type: array required: false - name: utterances type: array required: false - name: translation type: object required: false properties: - name: success type: boolean required: false - name: is_empty type: boolean required: false - name: exec_time type: number required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: 'null' required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: integer required: false - name: type type: string description: string required: false - name: description type: string description: HTTP-like code or provider-specific status string. required: false - name: exception type: string description: Short machine-readable error name. required: false - name: message type: string description: Human-readable error explanation. required: false - name: results type: array required: false - name: summarization type: object required: false properties: - name: success type: boolean required: false - name: is_empty type: boolean required: false - name: exec_time type: number required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: 'null' required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: integer required: false - name: type type: string description: string required: false - name: description type: string description: HTTP-like code or provider-specific status string. required: false - name: exception type: string description: Short machine-readable error name. required: false - name: message type: string description: Human-readable error explanation. required: false - name: results type: string required: false - name: named_entity_recognition type: object required: false properties: - name: success type: boolean required: false - name: is_empty type: boolean required: false - name: exec_time type: number required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: 'null' required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: integer required: false - name: type type: string description: string required: false - name: description type: string description: HTTP-like code or provider-specific status string. required: false - name: exception type: string description: Short machine-readable error name. required: false - name: message type: string description: Human-readable error explanation. required: false - name: entity type: string required: false - name: sentiment_analysis type: object required: false properties: - name: success type: boolean required: false - name: is_empty type: boolean required: false - name: exec_time type: number required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: 'null' required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: integer required: false - name: type type: string description: string required: false - name: description type: string description: HTTP-like code or provider-specific status string. required: false - name: exception type: string description: Short machine-readable error name. required: false - name: message type: string description: Human-readable error explanation. required: false - name: results type: string required: false - name: chapterization type: object required: false properties: - name: success type: boolean required: false - name: is_empty type: boolean required: false - name: exec_time type: number required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: 'null' required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: integer required: false - name: type type: string description: string required: false - name: description type: string description: HTTP-like code or provider-specific status string. required: false - name: exception type: string description: Short machine-readable error name. required: false - name: message type: string description: Human-readable error explanation. required: false - name: results type: object required: false headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - data properties: session_id: *ref_1 created_at: *ref_2 type: type: string const: post_final_transcript x-parser-schema-id: data: type: object properties: metadata: type: object properties: audio_duration: type: number x-parser-schema-id: number_of_distinct_channels: type: integer x-parser-schema-id: billing_time: type: number x-parser-schema-id: transcription_time: type: number x-parser-schema-id: x-parser-schema-id: transcription: *ref_6 translation: type: object properties: success: type: boolean x-parser-schema-id: is_empty: type: boolean x-parser-schema-id: exec_time: type: number x-parser-schema-id: error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: results: type: array items: type: object properties: error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: full_transcript: type: string x-parser-schema-id: languages: type: array items: type: string x-parser-schema-id: x-parser-schema-id: sentences: type: array items: *ref_7 x-parser-schema-id: subtitles: type: array items: *ref_8 x-parser-schema-id: utterances: type: array items: *ref_4 x-parser-schema-id: x-parser-schema-id: x-parser-schema-id: x-parser-schema-id: summarization: type: object properties: success: type: boolean x-parser-schema-id: is_empty: type: boolean x-parser-schema-id: exec_time: type: number x-parser-schema-id: error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: results: type: string x-parser-schema-id: x-parser-schema-id: named_entity_recognition: type: object properties: success: type: boolean x-parser-schema-id: is_empty: type: boolean x-parser-schema-id: exec_time: type: number x-parser-schema-id: error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: entity: type: string x-parser-schema-id: x-parser-schema-id: sentiment_analysis: type: object properties: success: type: boolean x-parser-schema-id: is_empty: type: boolean x-parser-schema-id: exec_time: type: number x-parser-schema-id: error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: results: type: string x-parser-schema-id: x-parser-schema-id: chapterization: type: object properties: success: type: boolean x-parser-schema-id: is_empty: type: boolean x-parser-schema-id: exec_time: type: number x-parser-schema-id: error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: results: type: object x-parser-schema-id: x-parser-schema-id: x-parser-schema-id: description: >- Final aggregated output for the session, including metadata and add-on results. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:45:00Z' type: post_final_transcript data: metadata: audio_duration: 123.45 number_of_distinct_channels: 1 billing_time: 120 transcription_time: 3.2 transcription: full_transcript: hello world x-parser-schema-id: PostFinalTranscriptMessage title: Final transcript (aggregated) description: >- This message contains the informations about the post final transcript of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:45:00Z", "type": "post_final_transcript", "data": { "metadata": { "audio_duration": 123.45, "number_of_distinct_channels": 1, "billing_time": 120, "transcription_time": 3.2 }, "transcription": { "full_transcript": "hello world" } } } bindings: [] extensions: - id: x-parser-unique-object-id value: postFinalTranscript bindings: [] extensions: *ref_0 - &ref_21 id: onPostChapterization title: On post chapterization type: send messages: - &ref_40 id: postChapterization contentType: application/json payload: - name: Chapterization description: >- This message contains the informations about the post chapterization of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: 'null' required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: integer required: false - name: type type: string description: string required: false - name: description type: string description: HTTP-like code or provider-specific status string. required: false - name: exception type: string description: Short machine-readable error name. required: false - name: message type: string description: Human-readable error explanation. required: false - name: type type: string description: post_chapterization required: true - name: data type: object required: true properties: - name: results type: array required: false headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - data properties: session_id: *ref_1 created_at: *ref_2 error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: type: type: string const: post_chapterization x-parser-schema-id: data: type: object properties: results: type: array items: type: object properties: abstractive_summary: type: string x-parser-schema-id: extractive_summary: type: string x-parser-schema-id: summary: type: string x-parser-schema-id: headline: type: string x-parser-schema-id: gist: type: string x-parser-schema-id: keywords: type: array items: type: string x-parser-schema-id: x-parser-schema-id: start: type: number x-parser-schema-id: end: type: number x-parser-schema-id: sentences: type: array items: type: object properties: sentence: type: string x-parser-schema-id: start: type: number x-parser-schema-id: end: type: number x-parser-schema-id: words: type: array items: *ref_9 x-parser-schema-id: x-parser-schema-id: x-parser-schema-id: text: type: string x-parser-schema-id: x-parser-schema-id: x-parser-schema-id: x-parser-schema-id: description: Post-processing output with chapter-level summaries and metadata. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:46:00Z' type: post_chapterization error: null data: results: - headline: Project kickoff start: 0 end: 60 keywords: - timeline - owner x-parser-schema-id: PostChapterizationMessage title: Chapterization description: >- This message contains the informations about the post chapterization of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:46:00Z", "type": "post_chapterization", "error": null, "data": { "results": [ { "headline": "Project kickoff", "start": 0, "end": 60, "keywords": [ "timeline", "owner" ] } ] } } bindings: [] extensions: - id: x-parser-unique-object-id value: postChapterization bindings: [] extensions: *ref_0 - &ref_22 id: onPostSummarization title: On post summarization type: send messages: - &ref_41 id: postSummarization contentType: application/json payload: - name: Summarization description: >- This message contains the informations about the post summarization of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: 'null' required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: integer required: false - name: type type: string description: string required: false - name: description type: string description: HTTP-like code or provider-specific status string. required: false - name: exception type: string description: Short machine-readable error name. required: false - name: message type: string description: Human-readable error explanation. required: false - name: type type: string description: post_summarization required: true - name: data type: object required: true properties: - name: results type: string required: false headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - data properties: session_id: *ref_1 created_at: *ref_2 error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: type: type: string const: post_summarization x-parser-schema-id: data: type: object properties: results: type: string x-parser-schema-id: x-parser-schema-id: description: Post-processing abstractive summary of the conversation. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:47:00Z' type: post_summarization error: null data: results: The team aligned on goals and next steps for the quarter. x-parser-schema-id: PostSummarizationMessage title: Summarization description: >- This message contains the informations about the post summarization of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:47:00Z", "type": "post_summarization", "error": null, "data": { "results": "The team aligned on goals and next steps for the quarter." } } bindings: [] extensions: - id: x-parser-unique-object-id value: postSummarization bindings: [] extensions: *ref_0 - &ref_23 id: onAudioChunkAck title: On audio chunk ack type: send messages: - &ref_42 id: audioChunkAck contentType: application/json payload: - name: Audio Chunk Acknowledgment description: >- This message contains the informations about the audio chunk acknowledgment of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: acknowledged type: boolean description: >- True if the server successfully received and buffered the audio chunk. required: true - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: 'null' required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: integer required: false - name: type type: string description: string required: false - name: description type: string description: HTTP-like code or provider-specific status string. required: false - name: exception type: string description: Short machine-readable error name. required: false - name: message type: string description: Human-readable error explanation. required: false - name: type type: string description: Event type for audio chunk acknowledgments. required: true - name: data type: object required: true properties: - name: byte_range type: array description: >- Inclusive byte offsets [start, end] acknowledged for this chunk. required: false - name: time_range type: array description: >- Inclusive audio time range in seconds [start, end] covered by this chunk. required: false headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - acknowledged - data properties: session_id: *ref_1 created_at: *ref_2 acknowledged: type: boolean description: >- True if the server successfully received and buffered the audio chunk. x-parser-schema-id: error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: type: type: string const: audio_chunk description: Event type for audio chunk acknowledgments. x-parser-schema-id: data: type: object properties: byte_range: type: array items: type: number x-parser-schema-id: minItems: 2 maxItems: 2 description: >- Inclusive byte offsets [start, end] acknowledged for this chunk. x-parser-schema-id: time_range: type: array items: type: number x-parser-schema-id: minItems: 2 maxItems: 2 description: >- Inclusive audio time range in seconds [start, end] covered by this chunk. x-parser-schema-id: x-parser-schema-id: description: >- Acknowledges receipt of an audio chunk and indicates the acknowledged byte/time ranges. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:34:56Z' type: audio_chunk acknowledged: true error: null data: byte_range: - 0 - 4095 time_range: - 0 - 0.256 x-parser-schema-id: AudioChunkAckMessage title: Audio Chunk Acknowledgment description: >- This message contains the informations about the audio chunk acknowledgment of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:34:56Z", "type": "audio_chunk", "acknowledged": true, "error": null, "data": { "byte_range": [ 0, 4095 ], "time_range": [ 0, 0.256 ] } } bindings: [] extensions: - id: x-parser-unique-object-id value: audioChunkAck bindings: [] extensions: *ref_0 - &ref_24 id: onStopRecordingAck title: On stop recording ack type: send messages: - &ref_43 id: stopRecordingAck contentType: application/json payload: - name: Stop Recording Acknowledgment description: >- This message contains the informations about the stop recording acknowledgment of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: acknowledged type: boolean description: True if stop_recording was accepted; false if rejected. required: true - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: 'null' required: false - name: oneOf type: oneOf description: Must be one of these types properties: - name: type type: string description: integer required: false - name: type type: string description: string required: false - name: description type: string description: HTTP-like code or provider-specific status string. required: false - name: exception type: string description: Short machine-readable error name. required: false - name: message type: string description: Human-readable error explanation. required: false - name: type type: string description: Event type for stop recording acknowledgments. required: true - name: data type: object description: >- Additional context provided by the provider; may be empty when acknowledged is true. required: true headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - acknowledged - data properties: session_id: *ref_1 created_at: *ref_2 acknowledged: type: boolean description: True if stop_recording was accepted; false if rejected. x-parser-schema-id: error: oneOf: - type: 'null' x-parser-schema-id: - *ref_5 x-parser-schema-id: type: type: string const: stop_recording description: Event type for stop recording acknowledgments. x-parser-schema-id: data: type: object description: >- Additional context provided by the provider; may be empty when acknowledged is true. x-parser-schema-id: description: >- Acknowledges the stop recording command with optional provider details. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:35:10Z' type: stop_recording acknowledged: true error: null data: {} - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:35:12Z' type: stop_recording acknowledged: false error: status_code: 409 exception: ConflictError message: Recording already stopped data: {} x-parser-schema-id: StopRecordingAckMessage title: Stop Recording Acknowledgment description: >- This message contains the informations about the stop recording acknowledgment of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:35:10Z", "type": "stop_recording", "acknowledged": true, "error": null, "data": {} } bindings: [] extensions: - id: x-parser-unique-object-id value: stopRecordingAck bindings: [] extensions: *ref_0 - &ref_25 id: onStartSession title: On start session type: send messages: - &ref_44 id: startSession contentType: application/json payload: - name: Start Session description: >- This message contains the informations about the start session of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: type type: string description: >- Lifecycle event emitted once when the session is created and ready. required: true headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type properties: session_id: *ref_1 created_at: *ref_2 type: type: string const: start_session description: >- Lifecycle event emitted once when the session is created and ready. x-parser-schema-id: description: Base lifecycle event for session start. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:34:00Z' type: start_session x-parser-schema-id: SimpleLifecycleMessage title: Start Session description: >- This message contains the informations about the start session of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:34:00Z", "type": "start_session" } bindings: [] extensions: - id: x-parser-unique-object-id value: startSession bindings: [] extensions: *ref_0 - &ref_26 id: onStartRecording title: On start recording type: send messages: - &ref_45 id: startRecording contentType: application/json payload: - name: Start Recording description: >- This message contains the informations about the start recording of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: type type: string description: Lifecycle event emitted when audio capture/streaming begins. required: true headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type properties: session_id: *ref_1 created_at: *ref_2 type: type: string const: start_recording description: Lifecycle event emitted when audio capture/streaming begins. x-parser-schema-id: description: Lifecycle event indicating the start of recording. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:34:05Z' type: start_recording x-parser-schema-id: SimpleLifecycleMessageStart title: Start Recording description: >- This message contains the informations about the start recording of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:34:05Z", "type": "start_recording" } bindings: [] extensions: - id: x-parser-unique-object-id value: startRecording bindings: [] extensions: *ref_0 - &ref_27 id: onEndRecording title: On end recording type: send messages: - &ref_46 id: endRecording contentType: application/json payload: - name: End Recording description: >- This message contains the informations about the end recording of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: type type: string description: >- Lifecycle event emitted when the provider has stopped receiving audio. required: true - name: data type: object description: >- Provider-defined object with optional metadata related to the end of recording. required: true headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type - data properties: session_id: *ref_1 created_at: *ref_2 type: type: string const: end_recording description: >- Lifecycle event emitted when the provider has stopped receiving audio. x-parser-schema-id: data: type: object additionalProperties: true description: >- Provider-defined object with optional metadata related to the end of recording. x-parser-schema-id: description: >- Lifecycle event sent when recording has ended, optionally carrying provider metadata. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:36:00Z' type: end_recording data: reason: user_request received_total_bytes: 1048576 x-parser-schema-id: EndRecordingMessage title: End Recording description: >- This message contains the informations about the end recording of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:36:00Z", "type": "end_recording", "data": { "reason": "user_request", "received_total_bytes": 1048576 } } bindings: [] extensions: - id: x-parser-unique-object-id value: endRecording bindings: [] extensions: *ref_0 - &ref_28 id: onEndSession title: On end session type: send messages: - &ref_47 id: endSession contentType: application/json payload: - name: End Session description: >- This message contains the informations about the end session of the audio chunk. type: object properties: - name: session_id type: string description: UUID identifying the session. required: true - name: created_at type: string description: ISO 8601 timestamp in UTC. required: true - name: type type: string description: >- Lifecycle event emitted when the session is closed and no further data will be sent. required: true headers: [] jsonPayloadSchema: type: object required: - session_id - created_at - type properties: session_id: *ref_1 created_at: *ref_2 type: type: string const: end_session description: >- Lifecycle event emitted when the session is closed and no further data will be sent. x-parser-schema-id: description: Lifecycle event for session termination. examples: - session_id: 550e8400-e29b-41d4-a716-446655440000 created_at: '2025-09-19T12:36:05Z' type: end_session x-parser-schema-id: SimpleLifecycleMessageEndSession title: End Session description: >- This message contains the informations about the end session of the audio chunk. example: |- { "session_id": "550e8400-e29b-41d4-a716-446655440000", "created_at": "2025-09-19T12:36:05Z", "type": "end_session" } bindings: [] extensions: - id: x-parser-unique-object-id value: endSession bindings: [] extensions: *ref_0 sendOperations: - *ref_10 - *ref_11 - *ref_12 receiveOperations: - *ref_13 - *ref_14 - *ref_15 - *ref_16 - *ref_17 - *ref_18 - *ref_19 - *ref_20 - *ref_21 - *ref_22 - *ref_23 - *ref_24 - *ref_25 - *ref_26 - *ref_27 - *ref_28 sendMessages: - *ref_29 - *ref_30 - *ref_31 receiveMessages: - *ref_32 - *ref_33 - *ref_34 - *ref_35 - *ref_36 - *ref_37 - *ref_38 - *ref_39 - *ref_40 - *ref_41 - *ref_42 - *ref_43 - *ref_44 - *ref_45 - *ref_46 - *ref_47 extensions: - id: x-parser-unique-object-id value: liveTranscription securitySchemes: [] ````