bzikst/higgs-audio-v3-tts-4b

Public
441 runs

Run bzikst/higgs-audio-v3-tts-4b with an API

Use one of our client libraries to get started quickly. Clicking on a library will take you to the Playground tab where you can tweak different inputs, see the results, and copy the corresponding code to use in your own project.

Input schema

The fields you can use to run this model with an API. If you don't give a value for a field its default value will be used.

Field Type Default value Description
text
string
Hello, this is Higgs Audio v3 TTS.
Text to synthesize. Supports inline Higgs control tokens.
sentence_pause_mode
None
off
Automatically insert Higgs pause tokens after sentence endings.
reference_audio
string
Optional reference audio for zero-shot voice cloning (WAV/MP3).
reference_text
string
Transcript of the reference audio; materially improves cloning quality.
voice
string
default
Preset voice name (ignored when reference_audio is set).
response_format
None
wav
Output audio format.
temperature
number
1

Max: 2

Sampling temperature.
top_p
number

Max: 1

Top-p (nucleus) sampling. Unset = server default.
top_k
integer
Top-k sampling. Unset = server default.
max_new_tokens
integer
2048

Min: 1

Max: 8192

Maximum number of generated multi-codebook steps.
seed
integer
Random seed for reproducibility. Unset = random.

Output schema

The shape of the response you’ll get when you run this model with an API.

Schema
{
  "type": "string",
  "title": "Output",
  "format": "uri"
}