HeyGen Voice is HeyGen’s in-house voice model. Use a public voice or create a clone from a recording. Public voices and clones use the same speech endpoints. An instant voice can also speak a recording again: the same words and timing, in that voice.
HeyGen Voice Turbo Free is free until November 30, 2026. Use
"model": "heygen-voice-1-turbo-free" with a public or instant voice. No credit balance is required. We will notify you before this offer changes.Try it in the playground · Get a free API keyChoose a model
Use Turbo Free for tests, prototypes, and small amounts of speech; its
429 and 500 limits and retry steps are on Turbo Free.
Hear the public voices
Listen to public voice samples in the playground, then use the same voice in the API. Start with Nadine (83548de556df48ba8c09c42a57c51d85) in the free speech example. No clone is needed.
To find more samples, call GET /v3/models/audio/voices?type=public and filter it by language, gender, age, or a free-text query. Each ACTIVE voice includes a preview_audio_url. Play that URL to hear the voice before you generate speech.
Choose a mode
Endpoints
Send your API key in the
X-Api-Key header; create one in the API keys page. Voices belong to the workspace that created them.
Voice slots
An instant voice is free to create during the preview and counts toward the workspace’s voice clone allowance. A professional voice takes one purchased slot from the Usage page; trainings per slot and thevoice_expired behaviour past the slot limit are on Slots and training allowance.
Next steps
The samevoice_id works in a video: pass it to POST /v3/videos. See Voices.
Instant Voice Clone
One recording, ready in seconds.
Professional Voice Clone
Train from 20+ minutes of recordings.
Text to Speech
Completed or streaming speech from any
ACTIVE voice.Speech to Speech
Speak a recording again in an instant voice.

