> ## 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. # Fish Audio TTS S2.1 Pro > Stream speech synthesis from Fish Speech S2.1 Pro over a native MessagePack WebSocket. Send a start frame followed by text frames; receive binary audio chunks. ## AsyncAPI ````yaml api-reference/asyncapi/tts-fish.asyncapi.yaml /v1/tts/fish/tts:s2.1-pro id: /v1/tts/fish/tts:s2.1-pro title: Fish Audio TTS S2.1 Pro description: >- Stream speech synthesis from Fish Speech S2.1 Pro over a native MessagePack WebSocket. Send a start frame followed by text frames; receive binary audio chunks. servers: - id: production protocol: wss host: api.slng.ai bindings: [] variables: [] - id: staging protocol: wss host: stageapi.slng.ai bindings: [] variables: [] address: /v1/tts/fish/tts:s2.1-pro parameters: [] bindings: - protocol: ws version: latest value: method: GET headers: type: object properties: X-World-Part-Override: type: string description: >- Target world part override. Auto-selected if not provided. Available world parts: `eu`. enum: - eu x-parser-schema-id: x-parser-schema-id: WebsocketHeadersFish schemaProperties: - name: method type: string description: GET required: false - name: headers type: object required: false properties: - name: X-World-Part-Override type: string description: >- Target world part override. Auto-selected if not provided. Available world parts: `eu`. enumValues: - eu required: false operations: - &ref_1 id: ttsFishTtsS21ProReceiveStart title: Tts fish tts s21 pro receive start description: Start a synthesis session on Fish Audio TTS S2.1 Pro type: receive messages: - &ref_8 id: FishTtsStartRequest contentType: application/msgpack payload: - name: Start Request description: >- Open a Fish Audio synthesis session with voice and audio configuration. type: object properties: - name: event type: string description: start required: true - name: request type: object description: >- Synthesis configuration for a Fish Audio WebSocket start frame. required: true properties: - name: text type: string description: >- Optional priming text. May be empty in a start frame; text is normally sent via subsequent text frames. required: false - name: reference_id type: string description: >- Fish Audio voice-model ID. For multi-speaker synthesis, provide one ID per speaker as an array. required: false - name: references type: array description: >- Inline reference audio for voice cloning (MessagePack binary values). required: false properties: - name: audio type: string description: Reference audio bytes. required: true - name: text type: string description: Transcript of the reference audio. required: true - name: prosody type: object description: Prosody controls. required: false properties: - name: speed type: number description: Speech speed multiplier. required: false - name: volume type: number description: Volume adjustment in dB. required: false - name: normalize_loudness type: boolean description: Whether to normalize output loudness. required: false - name: temperature type: number description: Sampling temperature. required: false - name: top_p type: number description: Nucleus sampling probability mass. required: false - name: chunk_length type: integer description: Target chunk length for synthesis. required: false - name: normalize type: boolean description: Whether to normalize input text before synthesis. required: false - name: format type: string description: Output audio format. enumValues: - wav - pcm - mp3 - opus required: false - name: sample_rate type: integer description: Output sample rate in Hz. enumValues: - 8000 - 16000 - 24000 - 32000 - 44100 - 48000 required: false - name: mp3_bitrate type: integer description: MP3 bitrate in kbps. enumValues: - 64 - 128 - 192 required: false - name: opus_bitrate type: integer description: Opus bitrate in bps. `-1000` selects the provider default. enumValues: - -1000 - 24000 - 32000 - 48000 - 64000 required: false - name: latency type: string description: Latency/quality trade-off. enumValues: - low - normal - balanced required: false - name: max_new_tokens type: integer description: Maximum number of new tokens to generate. required: false - name: repetition_penalty type: number description: Penalty applied to repeated tokens. required: false - name: min_chunk_length type: integer description: Minimum chunk length for synthesis. required: false - name: condition_on_previous_chunks type: boolean description: >- Whether generation is conditioned on previously generated chunks. required: false - name: early_stop_threshold type: number description: Early-stop threshold for generation. required: false - name: features type: array description: Optional feature flags. required: false properties: - name: item type: string required: false headers: [] jsonPayloadSchema: type: object description: >- MessagePack start frame. Must be the first message; opens the synthesis session. required: - event - request properties: event: type: string const: start x-parser-schema-id: request: type: object description: Synthesis configuration for a Fish Audio WebSocket start frame. properties: text: type: string description: >- Optional priming text. May be empty in a start frame; text is normally sent via subsequent text frames. x-parser-schema-id: reference_id: type: string description: >- Fish Audio voice-model ID. For multi-speaker synthesis, provide one ID per speaker as an array. x-parser-schema-id: references: type: array description: >- Inline reference audio for voice cloning (MessagePack binary values). items: type: object description: >- Inline reference audio for voice cloning, carried as MessagePack binary. required: - audio - text properties: audio: type: string format: binary description: Reference audio bytes. x-parser-schema-id: text: type: string description: Transcript of the reference audio. x-parser-schema-id: x-parser-schema-id: FishTtsReference x-parser-schema-id: prosody: type: object description: Prosody controls. properties: speed: type: number minimum: 0.5 maximum: 2 default: 1 description: Speech speed multiplier. x-parser-schema-id: volume: type: number minimum: -20 maximum: 20 default: 0 description: Volume adjustment in dB. x-parser-schema-id: normalize_loudness: type: boolean default: true description: Whether to normalize output loudness. x-parser-schema-id: x-parser-schema-id: temperature: type: number minimum: 0 maximum: 1 default: 0.7 description: Sampling temperature. x-parser-schema-id: top_p: type: number minimum: 0 maximum: 1 default: 0.7 description: Nucleus sampling probability mass. x-parser-schema-id: chunk_length: type: integer minimum: 100 maximum: 300 default: 300 description: Target chunk length for synthesis. x-parser-schema-id: normalize: type: boolean default: true description: Whether to normalize input text before synthesis. x-parser-schema-id: format: type: string enum: - wav - pcm - mp3 - opus default: mp3 description: Output audio format. x-parser-schema-id: sample_rate: type: integer nullable: true enum: - 8000 - 16000 - 24000 - 32000 - 44100 - 48000 description: Output sample rate in Hz. x-parser-schema-id: mp3_bitrate: type: integer enum: - 64 - 128 - 192 default: 128 description: MP3 bitrate in kbps. x-parser-schema-id: opus_bitrate: type: integer enum: - -1000 - 24000 - 32000 - 48000 - 64000 default: -1000 description: Opus bitrate in bps. `-1000` selects the provider default. x-parser-schema-id: latency: type: string enum: - low - normal - balanced default: normal description: Latency/quality trade-off. x-parser-schema-id: max_new_tokens: type: integer default: 1024 description: Maximum number of new tokens to generate. x-parser-schema-id: repetition_penalty: type: number default: 1.2 description: Penalty applied to repeated tokens. x-parser-schema-id: min_chunk_length: type: integer minimum: 0 maximum: 100 default: 50 description: Minimum chunk length for synthesis. x-parser-schema-id: condition_on_previous_chunks: type: boolean default: true description: >- Whether generation is conditioned on previously generated chunks. x-parser-schema-id: early_stop_threshold: type: number minimum: 0 maximum: 1 default: 1 description: Early-stop threshold for generation. x-parser-schema-id: features: type: array description: Optional feature flags. items: type: string x-parser-schema-id: x-parser-schema-id: x-parser-schema-id: FishTtsWsRequest x-parser-schema-id: title: Start Request description: >- Open a Fish Audio synthesis session with voice and audio configuration. example: |- { "event": "start", "request": { "reference_id": "9a9cf47702da476aa4629e2506d4a857", "format": "mp3" } } bindings: [] extensions: - id: x-parser-unique-object-id value: FishTtsStartRequest bindings: [] extensions: &ref_0 - id: x-parser-unique-object-id value: /v1/tts/fish/tts:s2.1-pro - &ref_2 id: ttsFishTtsS21ProReceiveText title: Tts fish tts s21 pro receive text description: Send text to Fish Audio TTS S2.1 Pro type: receive messages: - &ref_9 id: FishTtsTextMessage contentType: application/msgpack payload: - name: Text Frame description: Append text to synthesize on the active session. type: object properties: - name: event type: string description: text required: true - name: text type: string description: Text to synthesize. required: true headers: [] jsonPayloadSchema: type: object description: MessagePack text frame carrying text to synthesize. required: - event - text properties: event: type: string const: text x-parser-schema-id: text: type: string minLength: 1 description: Text to synthesize. x-parser-schema-id: x-parser-schema-id: title: Text Frame description: Append text to synthesize on the active session. example: |- { "event": "text", "text": "Hello from Fish Audio." } bindings: [] extensions: - id: x-parser-unique-object-id value: FishTtsTextMessage bindings: [] extensions: *ref_0 - &ref_3 id: ttsFishTtsS21ProReceiveFlush title: Tts fish tts s21 pro receive flush description: Flush buffered text on Fish Audio TTS S2.1 Pro type: receive messages: - &ref_10 id: FishTtsFlushMessage contentType: application/msgpack payload: - name: Flush Frame description: Force synthesis of buffered text without ending the session. type: object properties: - name: event type: string description: flush required: true headers: [] jsonPayloadSchema: type: object description: MessagePack flush frame. required: - event properties: event: type: string const: flush x-parser-schema-id: x-parser-schema-id: title: Flush Frame description: Force synthesis of buffered text without ending the session. example: |- { "event": "flush" } bindings: [] extensions: - id: x-parser-unique-object-id value: FishTtsFlushMessage bindings: [] extensions: *ref_0 - &ref_4 id: ttsFishTtsS21ProReceiveStop title: Tts fish tts s21 pro receive stop description: Stop the session on Fish Audio TTS S2.1 Pro type: receive messages: - &ref_11 id: FishTtsStopMessage contentType: application/msgpack payload: - name: Stop Frame description: End input; the server finishes synthesis and closes the session. type: object properties: - name: event type: string description: stop required: true headers: [] jsonPayloadSchema: type: object description: MessagePack stop frame. required: - event properties: event: type: string const: stop x-parser-schema-id: x-parser-schema-id: title: Stop Frame description: End input; the server finishes synthesis and closes the session. example: |- { "event": "stop" } bindings: [] extensions: - id: x-parser-unique-object-id value: FishTtsStopMessage bindings: [] extensions: *ref_0 - &ref_5 id: ttsFishTtsS21ProSendAudio title: Tts fish tts s21 pro send audio description: Receive an audio chunk from Fish Audio TTS S2.1 Pro type: send messages: - &ref_12 id: FishTtsAudioChunk contentType: application/msgpack payload: - name: Audio Chunk description: Streamed binary audio chunk from Fish Audio. type: object properties: - name: event type: string description: audio required: true - name: audio type: string description: Binary audio chunk (MessagePack binary value). required: true headers: [] jsonPayloadSchema: type: object description: >- MessagePack audio frame carrying a binary audio chunk in the requested format. required: - event - audio properties: event: type: string const: audio x-parser-schema-id: audio: type: string format: binary description: Binary audio chunk (MessagePack binary value). x-parser-schema-id: x-parser-schema-id: title: Audio Chunk description: Streamed binary audio chunk from Fish Audio. example: |- { "event": "audio", "audio": "" } bindings: [] extensions: - id: x-parser-unique-object-id value: FishTtsAudioChunk bindings: [] extensions: *ref_0 - &ref_6 id: ttsFishTtsS21ProSendFinish title: Tts fish tts s21 pro send finish description: Receive the finish signal from Fish Audio TTS S2.1 Pro type: send messages: - &ref_13 id: FishTtsFinish contentType: application/msgpack payload: - name: Finish description: End-of-synthesis signal for the session. type: object properties: - name: event type: string description: finish required: true - name: reason type: string description: Completion reason. enumValues: - stop - error required: true headers: [] jsonPayloadSchema: type: object description: >- MessagePack finish frame. `reason` indicates normal completion or an error stop. required: - event - reason properties: event: type: string const: finish x-parser-schema-id: reason: type: string enum: - stop - error description: Completion reason. x-parser-schema-id: x-parser-schema-id: title: Finish description: End-of-synthesis signal for the session. example: |- { "event": "finish", "reason": "stop" } bindings: [] extensions: - id: x-parser-unique-object-id value: FishTtsFinish bindings: [] extensions: *ref_0 - &ref_7 id: ttsFishTtsS21ProSendError title: Tts fish tts s21 pro send error description: Receive an error from Fish Audio TTS S2.1 Pro type: send messages: - &ref_14 id: FishTtsError contentType: application/msgpack payload: - name: Error description: Error frame from the provider. type: object properties: - name: event type: string description: error required: true - name: error type: string description: Human-readable error message. required: true headers: [] jsonPayloadSchema: type: object description: MessagePack error frame. required: - event - error properties: event: type: string const: error x-parser-schema-id: error: type: string minLength: 1 description: Human-readable error message. x-parser-schema-id: x-parser-schema-id: title: Error description: Error frame from the provider. example: |- { "event": "error", "error": "Synthesis failed" } bindings: [] extensions: - id: x-parser-unique-object-id value: FishTtsError bindings: [] extensions: *ref_0 sendOperations: - *ref_1 - *ref_2 - *ref_3 - *ref_4 receiveOperations: - *ref_5 - *ref_6 - *ref_7 sendMessages: - *ref_8 - *ref_9 - *ref_10 - *ref_11 receiveMessages: - *ref_12 - *ref_13 - *ref_14 extensions: - id: x-parser-unique-object-id value: /v1/tts/fish/tts:s2.1-pro securitySchemes: - id: bearer name: bearer type: http description: > API key issued by SLNG. Pass as `Authorization: Bearer ` in the WebSocket upgrade request headers. scheme: bearer extensions: [] ````