Echo-TTS

Jordan Darefsky, 2025. See technical details here. All audio outputs are subject to non-commercial use CC-BY-NC-SA-4.0. (This is a new, simpler Gradio UI; select "advanced" below to access the older version (with more hyperpameter control).

Upload a voice reference (or select a voice from the library), enter text (or select a text preset), and generate!

Generate up to 30 seconds of audio. If the generated voice does not match the reference speaker, enable Speaker KV in the Generation Parameters section.

Voice

Preset

Check this if the generated voice does NOT match the reference speaker


💡 For more control over generation parameters, switch to the Advanced tab.