TTS / STT

📄️Estimate speech cost

Estimate the cost of a speech request. No charge.

📄️Text to speech

Text to speech. Choose a character voice with slug and pass the text in text.

📄️Text to speech - OpenAI body

Text to speech with the OpenAI /audio/speech request body: voice and input.

📄️Transcribe audio

Speech to text. response_format is json, verbose_json, text, srt or vtt; timestamp_granularities[] is passed through. Billed by the length of the audio. Keys with a per-request cap cannot use it, because the amount is only known after processing.

📄️List voices

List voices. Requires a key.

📄️Entrust your own voice - alias

Alias of POST /v1/audio/voices.

📄️Revoke an entrusted voice - alias

Alias of DELETE /v1/audio/voices/{id}.

📄️Get an entrusted voice - alias

Alias of GET /v1/audio/voices/{id}.

📄️Voice verification statement - alias

Alias of POST /v1/audio/voices/challenges.

📄️Verification statement status - alias

Status of a verification statement (same as GET /v1/audio/voices/challenges/:id).

📄️Voice custody consent text - alias

The canonical voice custody consent text (same as GET /v1/audio/voices/consent).

📄️Japanese reading dictionary

Japanese reading dictionary: the replacements the server applies by default, longest term first, to Japanese text only.