Transcription
MOSS Transcribe Diarize
MOSS Transcribe DiarizeMOSS-Transcribe-Diarize is an end-to-end audio model for long-form multi-speaker transcription, diarization, timestamps, and acoustic event awareness.
Model details
View repositoryMOSS-Transcribe-Diarize 0.9B is an end-to-end audio understanding model for long-form multi-speaker transcription, diarization, timestamps, and acoustic event awareness.
It supports transcription and diarization across 50+ languages, single-pass inference on audio recordings up to 90 minutes long, and custom hotword prompting for domain-specific terms.
Input
1from pathlib import Path
2from openai import OpenAI
3
4model_id = "" # Place your Baseten model ID here.
5
6client = OpenAI(
7 api_key="BASETEN-API-KEY",
8 base_url=(
9 f"https://model-{model_id}.api.baseten.co/"
10 "environments/production/sync/v1"
11 ),
12 timeout=600.0,
13)
14
15with Path("audio.wav").open("rb") as audio_file:
16 response = client.audio.transcriptions.create(
17 model="OpenMOSS-Team/MOSS-Transcribe-Diarize",
18 file=audio_file,
19 response_format="verbose_json",
20 extra_body={"max_new_tokens": 65536},
21 )
22
23print(response.to_json())
