Voice cloning

sarvam-clone

Clone a voice from one clip. $30 per million characters.

The call is Pascal’s. Send Authorization: Bearer with a Pascal key, and Pascal’s body: model, the file or the text, and params. Inside params, a Pascal name and the provider’s own name are both accepted. If both are sent and they disagree, the request is rejected. The provider’s SDK and host are not this API. The API request fee is not charged. A studio call needs a Creator plan or an API balance.

Native waveform: One reference clip, at most 15 seconds, plus the text to speak.. Live price and availability are on GET /v1/models/sarvam-clone.

  • At most 1000 characters.
  • reference_text is optional and can help the clone.
  • Language defaults to English (India). This model does not detect it. en-IN is that same default.

Parameters

FieldTypeDescription
params.language
Default en-IN
stringThis model does not detect the language. Omitted uses English (India). en or en-IN is that same default. Bulbul’s languages, plus Assamese: en, hi, bn, gu, kn, ml, mr, pa, ta, te, as, or the -IN form. Odia is od or or. Both are the same. language_code is the same field.
params.voice_idstringA saved Sarvam voice. When this is set, the uploaded clip is not sent.

Example

curl https://api.pascalai.co/v1/voice-cloning \
  -H "Authorization: Bearer $PASCAL_API_KEY" \
  -H "Idempotency-Key: $(uuidgen)" \
  -F model=sarvam-clone \
  -F text="Keep walking toward the river." \
  -F params='{"language":"en-IN","output_format":"wav","output_sample_rate":"source","output_channels":"source"}' \
  -F file=@voice.wav

curl https://api.pascalai.co/v1/voice-cloning/$JOB_ID \
  -H "Authorization: Bearer $PASCAL_API_KEY"