Speech-driven Avatar
Send speech text, a Ruwana voice, speech rate, emotion and optional performance direction. Speech supports up to 3000 characters.
Ruwana Capability · Platform API live
Create a talking avatar from one image and speech.
Ruwana Avatar combines an avatar image with exactly one audio source: generated speech or uploaded finished audio. Platform measures prepared audio server-side, then runs a durable Standard 720p or Pro 1080p request.
Platform API liveText-to-speech or uploaded audio12 public voices7 emotionsTTS included
Uploaded-audio Avatar
Send your finished audio track instead of speech. TTS-only voice controls are not accepted when uploaded audio is used.
Public controls
12public Ruwana voices
7emotion presets
80–130%speech-rate range
3–30sprepared-audio production window
Voice names: Axel, Maya, Sienna, Ethan, Marcus, Elias, Nova, Celeste, Toby, Lila, Kai and Mimi. Emotions: Neutral, Confident, Happy, Angry, Sad, Presenter and Luxury.
| Tier | Resolution | Rate | TTS |
|---|---|---|---|
| Standard | 720p | $0.11/sec | Included |
| Pro | 1080p | $0.22/sec | Included |
Related Ruwana pages
Move from capability discovery to the production surface or implementation guide.
Avatar API
Product overview, controls and pricing.
Avatar docs
Speech/audio request rules and implementation examples.
Motion transfer
Use movement reference video instead of speech-driven performance.