vm6eji6m4/podcast-transcribe-zh

Chinese/Taiwanese podcast transcription with speaker diarization (Whisper large-v3-turbo + pyannote 3.1)

Public
21 runs

Run vm6eji6m4/podcast-transcribe-zh 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
audio
string
Audio file (mp3/wav/m4a/mp4), recommended < 60 min
language
None
zh
Language code. zh/ja/ko use large-v3-turbo, en uses distil-large-v3.
hotwords
string
Proper nouns to bias Whisper (e.g. '蔡康永 黃詹 OpenAI Anthropic').
enable_diarization
boolean
True
Run speaker diarization (requires hf_token). Disable to skip ~30% time.
gap_threshold
number
1.5

Min: 0.1

Max: 5

Merge adjacent same-speaker segments within this gap (seconds).
output_format
None
srt
Primary output format. JSON segments are always included.
hf_token
string
None

Output schema

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

Schema
{
  "type": "object",
  "title": "Output"
}