Audio

Text-to-speech

POST /v1/audio/speech

Returns raw audio bytes (Content-Type: audio/mpeg), not JSON.

Field	Type	Required	Description
`model`	string	required	TTS model id, e.g. `hexgrad/Kokoro-82M`.
`input`	string	required	Text to synthesize.
`voice`	string	default: af_bella	Voice id supported by the selected model.
`response_format`	string	default: mp3	`mp3` or `wav`.

curl https://api.aix.theaimart.co/v1/audio/speech \
  -H "Authorization: Bearer $AIX_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "hexgrad/Kokoro-82M", "input": "hello world", "voice": "af_bella" }' \
  --output speech.mp3

Billed by input tokens at the selected model’s published per-million-token rate.

Transcription

POST /v1/audio/transcriptions

Takes multipart form-data (not JSON).

Field	Type	Required	Description
`file`	file	required	Audio file (multipart form-data).
`model`	string	default: openai/whisper-large-v3-turbo	STT model id.
`prompt`	string	optional	Optional text that guides spelling, terminology or continuation context.
`response_format`	string	default: json	`json`, `text` or `verbose_json`. Non-verbose responses use the JSON text envelope.

curl https://api.aix.theaimart.co/v1/audio/transcriptions \
  -H "Authorization: Bearer $AIX_KEY" \
  -F file=@clip.mp3 \
  -F model=openai/whisper-large-v3-turbo

{ "text": "transcribed text here" }

Billed per minute using the measured transcript duration. AIx holds an estimate during the call and reconciles it to the returned duration. Maximum file size is 25 MB; accepted extensions are mp3, wav, m4a, flac, ogg and webm.

← Images

Models →