drama3-small

connecting…

Expressive Japanese / English speech on a single 12 GB GPU: generate a brand-new voice from a short description, or clone a preset voice. A compressed Fish Audio Drama 3 (9.7 GiB, IQ3 experts), running live on the owner's GPUs. Requests are queued; each GPU renders one at a time.

Voice

No reference audio: the model makes the voice up. Leave the description empty for a surprise. Every seed is a different voice. Like one? Press Keep this voice under the result to use it for more lines. Describe a kind of voice, not a real person.

Text

0 / 400
Expression tags

Point tags anywhere in the line: [laughs] [sighs] [gasps] [pause] [hesitates] [clears throat] [coughs]

Spans: <whisper>…</whisper> <emphasis>…</emphasis> <slow>…</slow>

Style (English, at the start): [with heavy, deflated sadness] [softly, whispering as if someone might overhear] [icy calm, barely containing anger]

Director's note: Direction: calm and reassuring. Slow down at the end. then a new line Line: …

Parameters

Terms of use

This is a free, non-commercial research demo. By generating audio you agree that you will not use this demo or its output to:

  • imitate the voice of an identifiable person without that person's prior consent;
  • impersonate any person or organization to deceive, defraud or harass, or place calls that break telemarketing or robocall laws;
  • create sexual content involving a minor or a voice presented as a minor's, or using a real person's voice without consent;
  • remove, disable or obscure any watermark or provenance signal, or train, distill or improve other speech or voice generation models.

Output is synthetic speech and provided as is. Requests (text, settings, IP address) are kept for up to 6 hours to run the queue and prevent abuse.

Recent