CastReader Voice APIMULTILINGUAL VOICE CLONING API
Your voice. Ready for another language.
Prepare an authorized recording once. Reuse its voice ID to turn your text into speech—without recording every line again.
Record · upload · invite the speaker
voice_idReuse for new textWorkflow illustration. This does not translate your text.
FROM RECORDING TO API
Clone once. Generate as often as you need.
Start with a permitted recording
Use your own voice, upload a permitted file or send a recording invitation from the console. Keep the recording clear and the transcript accurate.
Wait until the voice is ready
Registration prepares the reference for synthesis. Store the returned voice ID; switching between prepared voices does not require another registration.
Send text in the target language
Reuse the voice ID with text and its language. The SDK handles routing and queued-job waiting, then returns the generated audio.
import { VoiceAPI } from '@castreader/voice-api';
const client = new VoiceAPI();
// Use the ID of your ready, authorized clone.
const voiceId = 'YOUR_VOICE_ID';
const { audio } = await client.generate({
voice: voiceId,
text: '每一个故事，都值得被听见。',
language: 'zh',
output_format: 'mp3',
});
// Same voice ID. Different input text.
const english = await client.generate({
voice: voiceId,
text: 'Every story deserves to be heard.',
language: 'en',
output_format: 'mp3',
});Install the SDK and set CASTREADER_API_KEY first. Each successful generation is metered separately.
Registration fields and complete quickstart →PAY FOR GENERATION, NOT A SUBSCRIPTION
$8 / million characters.
One usage price for supported synthesis languages. Available trial characters are used first; waiting and downloading a retained result do not add another generation charge.
Know the boundaries before integrating.
Does voice cloning translate my content?
No. Provide Chinese text for Chinese speech, Japanese text for Japanese speech, and so on. Automatic detection can be uncertain for short or mixed-language input; specify the language when necessary. Pronunciation and speaker similarity vary by recording and language.
Can I build long audiobooks or live conversation?
Current queued tasks accept up to 500 characters, with 1 concurrent generation per workspace. Longer workflows need application-side splitting and audio assembly. This release does not expose streaming, real-time dialogue or emotion controls.
Who can access a private clone?
Private clones belong to their workspace and require authorized API access. They are not added to the public voice gallery. You must have permission from the speaker for cloning and the intended use.
Where do I start?
API access is currently by invitation. After activation, use the console to prepare a voice, try a short generation and copy the integration example. Reading-app subscriptions and API billing are separate.