We use cookies to enhance your experience, provide personalised content, and analyse our traffic; see our Cookie Policy.
fish-audio/fish-drama-3Clone a voice from one clip, 23.157 credits per 1,000 characters
Fish Audio's expressive speech model. Send one short audio clip of the voice you want in input_references and it reads your text in that voice, with no stored voice and no voice id. Direction goes in the text, in square brackets before a line, such as [nervous, whispering]. Billed per character: 23.157 credits per 1,000.
Last updated Nov 19, 2026
One audio clip is the voice: send it with each request, nothing is stored
Write the tone in square brackets before a line, such as [nervous, whispering]
You are billed for the characters you send, not for the audio length or the clip
23.157 credits per 1,000 characters
Yes. Top up your Tchavi account with Orange Money, MTN MoMo, M-Pesa or another mobile money operator. No bank card is needed. Credits land as soon as the payment is confirmed and are deducted as you use them.
Yes. Keep your current OpenAI SDK, just change the base URL to Tchavi's (shown in the API tab) and use your Tchavi key. The rest of your code stays the same.
import Tchavi from '@tchavi/sdk';
import fs from 'fs';
const client = new Tchavi({ apiKey: 'API_KEY' });
const response = await client.audio.speech.create({
"model": "fish-drama-3",
"input": "Hello, world!",
"voice": "alloy",
"response_format": "mp3"
});
const buffer = Buffer.from(await response.arrayBuffer());
fs.writeFileSync('speech.mp3', buffer);
console.log('X-Credits-Used:', response.headers.get('x-credits-used'));Every parameter this model accepts on the API. Check the SDK for full typings.
modelRequiredTTS model (e.g. `fish-drama-3`).
inputRequiredThe text to synthesize.
voiceRequiredVoice timbre: `alloy`, `ash`, `ballad`, `cedar`, `coral`, `echo`, `fable`, `marin`, `nova`, `onyx`, `sage`, or `shimmer`.
response_format"mp3"Audio container format.