Speech

Speech reference

TTS and transcription with provider rotation.

Endpoints

Method	Path	Scope	Description
POST	/v1/speech/tts	speech:tts	Text-to-speech synthesis.
POST	/v1/speech/transcribe	speech:stt	Transcribe an audio URL.
GET	/v1/speech/voices	speech:read	List available voices, grouped by language.

Quickstart

Try in playground →

curl -X POST https://api.valienz.io/v1/speech/tts \
  -H "Authorization: Bearer $VALIENZ_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "speech-tts-default",
    "voice": "rachel",
    "text": "Hello from Valienz"
  }'

Notes

Audio formats
STT accepts mp3, m4a, wav, flac, ogg, webm. Upload directly to MinIO via the presigned URL flow before submitting the transcription job.

Errors

Code	HTTP	Description	Resolution
E_UNSUPPORTED_AUDIO_FORMAT	400	Upload format not in the allow list.	Convert to one of mp3/m4a/wav/flac/ogg/webm before uploading.