← Home · Clone voices (EN/ES) and Hindi preset voices via Sarvam AI
1. Upload Audio
Record or upload a clear audio sample of the voice you want to clone. 10-20 seconds of continuous speech works best.
2. Clone
The audio is uploaded to cloud storage, then sent to DashScope for voice enrollment. This creates a unique voice profile.
3. Synthesize
Switch to the Avatars tab, find your cloned voice, type any text, and generate speech that sounds like the original speaker.