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 model that transcribes speech, labels speakers, and adds timestamps in a single pass, turning long, multi-speaker recordings into structured "who said what."
It supports 50+ languages, single-pass inference on recordings up to 90 minutes, acoustic-event awareness, and custom hotword prompting for domain-specific terms. It serves the OpenAI-compatible /v1/audio/transcriptions endpoint, and diarization and timestamps are on by default. Request verbose_json to get parsed speaker segments.
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())
24# Speakers are inline [S01], [S02], ... tags at the start of each segment's text.JSON output
1{
2 "duration": 30,
3 "text": "[6.15][S01]So guys, we need to come up with an idea to create a car that is revolutionary.[11.81][11.85][S01]Now, personally, the one thing I hate about cars is sitting in traffic.[15.91][16.16][S01]So how is our car going to avoid traffic completely? What do you guys think? Do you have any ideas?[21.45][18.08][S02]嗯。[18.51][20.63][S03]Well, have you ever seen Back to the Future?[22.34]...",
4 "segments": [
5 {
6 "id": 0,
7 "start": 6.15,
8 "end": 11.81,
9 "text": "[S01]So guys, we need to come up with an idea to create a car that is revolutionary."
10 },
11 {
12 "id": 1,
13 "start": 11.85,
14 "end": 15.91,
15 "text": "[S01]Now, personally, the one thing I hate about cars is sitting in traffic."
16 },
17 {
18 "id": 2,
19 "start": 16.16,
20 "end": 21.45,
21 "text": "[S01]So how is our car going to avoid traffic completely? What do you guys think? Do you have any ideas?"
22 },
23 {
24 "id": 3,
25 "start": 18.08,
26 "end": 18.51,
27 "text": "[S02]嗯。"
28 },
29 {
30 "id": 4,
31 "start": 20.63,
32 "end": 22.34,
33 "text": "[S03]Well, have you ever seen Back to the Future?"
34 },
35 {
36 "id": 5,
37 "start": 22.43,
38 "end": 25.21,
39 "text": "[S01]Yes, one of my best movies, I've seen it like three times."
40 },
41 {
42 "id": 6,
43 "start": 22.91,
44 "end": 23.34,
45 "text": "[S03]Where uh。"
46 },
47 {
48 "id": 7,
49 "start": 25.08,
50 "end": 26.75,
51 "text": "[S02]I was thinking like no wheels."
52 },
53 {
54 "id": 8,
55 "start": 27.17,
56 "end": 28.88,
57 "text": "[S01]No wheels, so how do we move around?"
58 },
59 {
60 "id": 9,
61 "start": 27.21,
62 "end": 27.86,
63 "text": "[S02]No wheels."
64 },
65 {
66 "id": 10,
67 "start": 28.91,
68 "end": 29.99,
69 "text": "[S02]Well, we're the only car that."
70 }
71 ],
72 "usage": {
73 "seconds": 30,
74 "type": "duration"
75 },
76 "task": "transcribe"
77}
