> ## Documentation Index > Fetch the complete documentation index at: https://docs.slng.ai/llms.txt > Use this file to discover all available pages before exploring further. # Soniox TTS v1 > Stream real-time speech synthesis from Soniox TTS v1 over WebSocket with low-latency incremental audio output and SLNG's unified TTS protocol. ## AsyncAPI ````yaml api-reference/asyncapi/tts-soniox.asyncapi.yaml /v1/tts/soniox/tts-rt:v1 id: /v1/tts/soniox/tts-rt:v1 title: Soniox TTS v1 description: >- Stream real-time speech synthesis from Soniox TTS v1 over WebSocket with low-latency incremental audio output and SLNG's unified TTS protocol. servers: - id: production protocol: wss host: api.slng.ai bindings: [] variables: [] - id: staging protocol: wss host: stageapi.slng.ai bindings: [] variables: [] address: /v1/tts/soniox/tts-rt:v1 parameters: [] bindings: - protocol: ws version: latest value: method: GET headers: type: object properties: X-World-Part-Override: type: string description: Override target world part. Auto-selected if not provided. enum: - na x-parser-schema-id: x-parser-schema-id: WebsocketHeadersSoniox schemaProperties: - name: method type: string description: GET required: false - name: headers type: object required: false properties: - name: X-World-Part-Override type: string description: Override target world part. Auto-selected if not provided. enumValues: - na required: false operations: - &ref_1 id: ttsSonioxTtsRtV1ReceiveConfig title: Tts soniox tts rt v1 receive config description: Initialize a TTS stream type: receive messages: - &ref_7 id: SonioxTtsRtV1Config contentType: application/json payload: - name: Connection Config Message description: >- Initialize a TTS stream. Sent once per stream_id before any text frames. type: object properties: - name: type type: string description: >- Message type discriminator. Always `init` for the connection config message. required: true - name: api_key type: string description: >- Soniox API key or temporary API key. Created in Soniox Console. Optional on gateway routes — when provided it is ignored; upstream auth is injected by the gateway. required: false - name: stream_id type: string description: >- Client-generated unique stream identifier. Must be unique among active streams on the same connection. Can be reused only after the previous stream is terminated. required: true - name: model type: string description: TTS model to use. Example, `tts-rt-v1`. required: true - name: language type: string description: ISO 639-1 language code. Example, `en`. enumValues: - af - sq - ar - az - eu - be - bn - bs - bg - ca - zh - hr - cs - da - nl - en - et - fi - fr - gl - de - el - gu - he - hi - hu - id - it - ja - kn - kk - ko - lv - lt - mk - ms - ml - mr - 'no' - fa - pl - pt - pa - ro - ru - sr - sk - sl - es - sw - sv - tl - ta - te - th - tr - uk - ur - vi - cy required: true - name: voice type: string description: Voice name. Example, `Adrian`. required: true - name: audio_format type: string description: Output audio format. Example, `wav`. enumValues: - wav - mp3 - pcm_s16le - pcm_s16be required: true - name: sample_rate type: integer description: Output sample rate in Hz. Example, `24000`. required: false - name: bitrate type: integer description: >- Codec bitrate in bps. Only applicable to lossy compressed formats. Example, `128000`. required: false headers: [] jsonPayloadSchema: type: object description: >- Initialize a TTS stream. Sent once per stream_id before any text frames. required: - type - stream_id - model - language - voice - audio_format properties: type: type: string const: init description: >- Message type discriminator. Always `init` for the connection config message. x-parser-schema-id: api_key: type: string description: >- Soniox API key or temporary API key. Created in Soniox Console. Optional on gateway routes — when provided it is ignored; upstream auth is injected by the gateway. x-parser-schema-id: stream_id: type: string description: >- Client-generated unique stream identifier. Must be unique among active streams on the same connection. Can be reused only after the previous stream is terminated. x-parser-schema-id: model: type: string default: tts-rt-v1 description: TTS model to use. Example, `tts-rt-v1`. x-parser-schema-id: language: type: string default: en description: ISO 639-1 language code. Example, `en`. enum: - af - sq - ar - az - eu - be - bn - bs - bg - ca - zh - hr - cs - da - nl - en - et - fi - fr - gl - de - el - gu - he - hi - hu - id - it - ja - kn - kk - ko - lv - lt - mk - ms - ml - mr - 'no' - fa - pl - pt - pa - ro - ru - sr - sk - sl - es - sw - sv - tl - ta - te - th - tr - uk - ur - vi - cy x-parser-schema-id: voice: type: string description: Voice name. Example, `Adrian`. x-parser-schema-id: audio_format: type: string enum: - wav - mp3 - pcm_s16le - pcm_s16be description: Output audio format. Example, `wav`. x-parser-schema-id: sample_rate: type: integer description: Output sample rate in Hz. Example, `24000`. x-parser-schema-id: bitrate: type: integer description: >- Codec bitrate in bps. Only applicable to lossy compressed formats. Example, `128000`. x-parser-schema-id: x-parser-schema-id: title: Connection Config Message description: >- Initialize a TTS stream. Sent once per stream_id before any text frames. example: |- { "type": "init", "stream_id": "stream-1", "model": "tts-rt-v1", "language": "en", "voice": "Adrian", "audio_format": "wav", "sample_rate": 24000 } bindings: [] extensions: - id: x-parser-unique-object-id value: SonioxTtsRtV1Config bindings: [] extensions: &ref_0 - id: x-parser-unique-object-id value: /v1/tts/soniox/tts-rt:v1 - &ref_2 id: ttsSonioxTtsRtV1ReceiveTextStreaming title: Tts soniox tts rt v1 receive text streaming description: Stream text to synthesize type: receive messages: - &ref_8 id: SonioxTtsRtV1TextStreaming contentType: application/json payload: - name: Text Streaming Message description: >- Stream text incrementally to the active stream. Set text_end true on the final chunk. type: object properties: - name: type type: string description: >- Message type discriminator. Always `text` for streaming text chunks. required: true - name: stream_id type: string description: >- Must match the stream_id used in the connection config message. required: true - name: text type: string description: >- Text chunk to synthesize. Send incrementally as text is generated. required: true - name: text_end type: boolean description: >- Set to true on the final text chunk to signal end of input. False for intermediate chunks. required: true headers: [] jsonPayloadSchema: type: object description: >- Stream text incrementally to the active stream. Set text_end true on the final chunk. required: - type - stream_id - text - text_end properties: type: type: string const: text description: >- Message type discriminator. Always `text` for streaming text chunks. x-parser-schema-id: stream_id: type: string description: Must match the stream_id used in the connection config message. x-parser-schema-id: text: type: string description: >- Text chunk to synthesize. Send incrementally as text is generated. x-parser-schema-id: text_end: type: boolean description: >- Set to true on the final text chunk to signal end of input. False for intermediate chunks. x-parser-schema-id: x-parser-schema-id: title: Text Streaming Message description: >- Stream text incrementally to the active stream. Set text_end true on the final chunk. example: |- { "type": "text", "stream_id": "stream-1", "text": "Hello from Soniox text-to-speech.", "text_end": false } bindings: [] extensions: - id: x-parser-unique-object-id value: SonioxTtsRtV1TextStreaming bindings: [] extensions: *ref_0 - &ref_3 id: ttsSonioxTtsRtV1ReceiveCancel title: Tts soniox tts rt v1 receive cancel description: Cancel an in-flight stream type: receive messages: - &ref_9 id: SonioxTtsRtV1Cancel contentType: application/json payload: - name: Cancel Message description: >- Cancel an in-flight stream. Server stops audio and terminates the stream. type: object properties: - name: type type: string description: >- Message type discriminator. Always `clear` for the cancel message. required: true - name: stream_id type: string description: Stream to cancel. required: true - name: const type: boolean required: false - name: description type: string description: >- Must be `true` to trigger cancellation. Server will not send further audio and will terminate the stream. required: false headers: [] jsonPayloadSchema: type: object description: >- Cancel an in-flight stream. Server stops audio and terminates the stream. required: - type - stream_id - cancel properties: type: type: string const: clear description: >- Message type discriminator. Always `clear` for the cancel message. x-parser-schema-id: stream_id: type: string description: Stream to cancel. x-parser-schema-id: cancel: const: true description: >- Must be `true` to trigger cancellation. Server will not send further audio and will terminate the stream. x-parser-schema-id: x-parser-schema-id: title: Cancel Message description: >- Cancel an in-flight stream. Server stops audio and terminates the stream. example: |- { "type": "clear", "stream_id": "stream-1", "cancel": true } bindings: [] extensions: - id: x-parser-unique-object-id value: SonioxTtsRtV1Cancel bindings: [] extensions: *ref_0 - &ref_4 id: ttsSonioxTtsRtV1SendAudioChunk title: Tts soniox tts rt v1 send audio chunk description: Receive synthesized audio chunks type: send messages: - &ref_10 id: SonioxTtsRtV1AudioChunk contentType: application/json payload: - name: Audio Response description: >- Audio bytes for the stream chunk. audio_end true marks the final audio payload (stream not yet terminated). type: object properties: - name: type type: string description: >- Message type discriminator. Always `audio_chunk` for audio response messages. required: true - name: audio type: string description: Base64-encoded audio bytes for the stream chunk. required: true - name: audio_end type: boolean description: >- True on the final audio chunk. No further audio payloads will follow, but the stream is not fully closed until a Terminal Stream Event arrives with terminated: true. required: true - name: stream_id type: string description: Identifies which stream this audio belongs to. required: true headers: [] jsonPayloadSchema: type: object description: >- Audio bytes for the stream chunk. audio_end true marks the final audio payload (stream not yet terminated). required: - type - audio - audio_end - stream_id properties: type: type: string const: audio_chunk description: >- Message type discriminator. Always `audio_chunk` for audio response messages. x-parser-schema-id: audio: type: string format: base64 description: Base64-encoded audio bytes for the stream chunk. x-parser-schema-id: audio_end: type: boolean description: >- True on the final audio chunk. No further audio payloads will follow, but the stream is not fully closed until a Terminal Stream Event arrives with terminated: true. x-parser-schema-id: stream_id: type: string description: Identifies which stream this audio belongs to. x-parser-schema-id: x-parser-schema-id: title: Audio Response description: >- Audio bytes for the stream chunk. audio_end true marks the final audio payload (stream not yet terminated). example: |- { "type": "audio_chunk", "audio": "UklGRiQAAABXQVZFZm10IBAAAAABAAEAQB8AAEAfAAABAAgAZGF0YQAAAAA=", "audio_end": false, "stream_id": "stream-1" } bindings: [] extensions: - id: x-parser-unique-object-id value: SonioxTtsRtV1AudioChunk bindings: [] extensions: *ref_0 - &ref_5 id: ttsSonioxTtsRtV1SendTerminated title: Tts soniox tts rt v1 send terminated description: Receive stream termination event type: send messages: - &ref_11 id: SonioxTtsRtV1Terminated contentType: application/json payload: - name: Terminal Stream Event description: >- Server has fully closed the stream and released resources. stream_id may now be reused. type: object properties: - name: type type: string description: >- Message type discriminator. Always `terminated` for the terminal stream event. required: true - name: const type: boolean required: false - name: description type: string description: >- True signals the server has fully closed the stream and released resources. Safe to reuse stream_id only after this event. required: false - name: stream_id type: string description: Stream this termination event applies to. required: true headers: [] jsonPayloadSchema: type: object description: >- Server has fully closed the stream and released resources. stream_id may now be reused. required: - type - terminated - stream_id properties: type: type: string const: terminated description: >- Message type discriminator. Always `terminated` for the terminal stream event. x-parser-schema-id: terminated: const: true description: >- True signals the server has fully closed the stream and released resources. Safe to reuse stream_id only after this event. x-parser-schema-id: stream_id: type: string description: Stream this termination event applies to. x-parser-schema-id: x-parser-schema-id: title: Terminal Stream Event description: >- Server has fully closed the stream and released resources. stream_id may now be reused. example: |- { "type": "terminated", "terminated": true, "stream_id": "stream-1" } bindings: [] extensions: - id: x-parser-unique-object-id value: SonioxTtsRtV1Terminated bindings: [] extensions: *ref_0 - &ref_6 id: ttsSonioxTtsRtV1SendError title: Tts soniox tts rt v1 send error description: Receive stream error type: send messages: - &ref_12 id: SonioxTtsRtV1Error contentType: application/json payload: - name: Error Response description: >- An error encountered on the stream. After an error the stream is typically terminated. type: object properties: - name: type type: string description: >- Message type discriminator. Always `error` for error responses. required: true - name: error_code type: integer description: >- HTTP-style error code. Common values: 400, 401, 402, 408, 429, 500, 503. required: true - name: error_message type: string description: >- Human-readable error description. Example: "Missing required field: model". required: true - name: stream_id type: string description: Stream that encountered the error. required: true headers: [] jsonPayloadSchema: type: object description: >- An error encountered on the stream. After an error the stream is typically terminated. required: - type - error_code - error_message - stream_id properties: type: type: string const: error description: Message type discriminator. Always `error` for error responses. x-parser-schema-id: error_code: type: integer description: >- HTTP-style error code. Common values: 400, 401, 402, 408, 429, 500, 503. x-parser-schema-id: error_message: type: string description: >- Human-readable error description. Example: "Missing required field: model". x-parser-schema-id: stream_id: type: string description: Stream that encountered the error. x-parser-schema-id: x-parser-schema-id: title: Error Response description: >- An error encountered on the stream. After an error the stream is typically terminated. example: |- { "type": "error", "error_code": 400, "error_message": "Missing required field: model", "stream_id": "stream-1" } bindings: [] extensions: - id: x-parser-unique-object-id value: SonioxTtsRtV1Error bindings: [] extensions: *ref_0 sendOperations: - *ref_1 - *ref_2 - *ref_3 receiveOperations: - *ref_4 - *ref_5 - *ref_6 sendMessages: - *ref_7 - *ref_8 - *ref_9 receiveMessages: - *ref_10 - *ref_11 - *ref_12 extensions: - id: x-parser-unique-object-id value: /v1/tts/soniox/tts-rt:v1 securitySchemes: - id: bearer name: bearer type: http scheme: bearer extensions: [] ````