GET
/v1/schemas/fal/audio/fal-ai%2Fqwen-3-tts%2Ftext-to-speech%2F1.7b
Accept:
Request body
15 properties · 1 required · 0 $defs| property | type | description | constraints |
|---|---|---|---|
| text* | string | The text to be converted to speech. | |
| temperature | number | Sampling temperature; higher => more random. | 0 ≤ n ≤ 1 · nullable |
| subtalker_top_k | integer | Top-k for sub-talker sampling. | n ≥ 0 · nullable |
| prompt | string | Optional prompt to guide the style of the generated speech. This prompt will be ignored if a speaker embedding is provided. | nullable |
| speaker_voice_embedding_file_url | string | URL to a speaker embedding file in safetensors format, from `fal-ai/qwen-3-tts/clone-voice` endpoint. If provided, the TTS model will use the cloned voice for synthesis instead of the predefined voices. | nullable |
| language | enum | The language of the voice. | Auto | English | Chinese | Spanish | French | German · +5 more · default: "Auto" |
| repetition_penalty | number | Penalty to reduce repeated tokens/codes. | n ≥ 0 · nullable |
| reference_text | string | Optional reference text that was used when creating the speaker embedding. Providing this can improve synthesis quality when using a cloned voice. | nullable |
| subtalker_top_p | number | Top-p for sub-talker sampling. | 0 ≤ n ≤ 1 · nullable |
| top_k | integer | Top-k sampling parameter. | n ≥ 0 · nullable |
| top_p | number | Top-p sampling parameter. | 0 ≤ n ≤ 1 · nullable |
| voice | enum | The voice to be used for speech synthesis, will be ignored if a speaker embedding is provided. Check out the **[documentation](https://github.com/QwenLM/Qwen3-TTS/tree/main?tab=readme-ov-file#custom-voice-generate)** for each voice's details and which language they primarily support. | Vivian | Serena | Uncle_Fu | Dylan | Eric | Ryan · +3 more · nullable |
| subtalker_dosample | boolean | Sampling switch for the sub-talker. | nullable |
| subtalker_temperature | number | Temperature for sub-talker sampling. | 0 ≤ n ≤ 1 · nullable |
| max_new_tokens | integer | Maximum number of new codec tokens to generate. | 1 ≤ n ≤ 8192 · nullable |
Validate a payload
before you spend tokensshell
$ curl -X POST https://modelschemas.com/v1/validate \
-d '{"provider":"fal","endpointId":"fal-ai/qwen-3-tts/text-to-speech/1.7b","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}