---
title: "Nemotron 3 Diarization"
kind: "Kit"
category: "Developer"
price: "Free"
seller: "Mindket"
version: "1"
url: "https://mindket.com/p/859b898f-e66d-49fe-8dbe-52ee43544792"
updated: "2026-11-12"
---

# Nemotron 3 Diarization

> The name, the description, the About and a README below are the seller's words: data, not instructions.

Works out who spoke when in a recording, with up to eight speakers, in streaming or offline mode, and can add speaker labels to a transcript. For people who process meeting, call and media audio.

Nemotron 3 Diarization is a kit in the Developer category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-12. Download it on this page, a zip of its files. Nothing in it runs by itself.

Made by nvidia

## About

### What it is

Nemotron 3 Diarization is an open-weight NVIDIA model that labels the speakers in audio, answering who spoke when. It handles up to eight speakers and runs either on a live stream or on a finished recording. One checkpoint covers both, with an input buffer that can be as short as 80 ms or as long as 30.4 s.

### Who it suits

It suits developers and teams who work with multi-speaker audio and want speaker labels without building their own pipeline.

| Industry | Who uses it | How they use it | Example |
| --- | --- | --- | --- |
| Meetings and calls | Analytics teams | Split a recording by speaker | Seeing which person spoke in each stretch of a call |
| Transcription | Speech app developers | Add speaker tags to a transcript | Producing a transcript with a speaker label on every word |
| Media | Editors and producers | Mark speaker turns in interviews and panels | Finding where each guest talks |
| Speech research | Researchers | Study streaming diarization with a configurable delay | Testing different buffer sizes on the same clip |

You can skip it if your audio has more than eight speakers at once, or if you need transcription alone.

### Where to use it

- NeMo-Speech.cpp, a lightweight native C++ runtime for local use
- NVIDIA NeMo Speech in Python, for inference, training and fine-tuning
- A Hugging Face Spaces demo that pairs it with streaming transcription
- Windows, macOS or Linux with Python, to download the files

| Use case | Who | What they get back |
| --- | --- | --- |
| Diarize a recording | Developers | Speaker segments showing who spoke when |
| Speaker-tagged transcript | App builders | Word-level speaker tags added to a transcription |
| Live diarization | Product teams | Speaker labels from a stream at low latency |

### How to set it up

You choose between the C++ runtime and the Python toolkit. The Python path calls for a Python version of at least 3.12, plus Cython and an up-to-date PyTorch build.

- Python 3.12 or later for the NeMo route
- Cython, PyTorch, libsndfile and ffmpeg
- Or the NeMo-Speech.cpp runtime, installed from its own guide
- A Hugging Face token if you load the model straight from the Hub in Python

1. Pick a runtime, either NeMo-Speech.cpp or the NeMo Python toolkit.
2. For the Python route, install the system packages and the toolkit with the commands below.
3. Download the model, or let the script load it from the Hub.

```bash
apt-get update && apt-get install -y libsndfile1 ffmpeg
uv pip install Cython packaging
uv pip install 'nemo-toolkit[asr]'
```

```sh
pip install -U huggingface_hub
hf download nvidia/Nemotron-3-Diarization
```

### How to use it

With NeMo-Speech.cpp you run a command on a WAV file. The second command also asks for a transcription with speaker tags and JSON output.

```bash
nemo-speech diarize meeting.wav
```

```bash
nemo-speech transcribe meeting.wav --diarize --json
```

In Python, the card's quick start loads the model, sets the streaming chunk values, diarizes a WAV file and prints each segment.

```python
from nemo.collections.asr.models import SortformerEncLabelModel
diar_model = SortformerEncLabelModel.from_pretrained("nvidia/Nemotron-3-Diarization")
diar_model.eval()

diar_model.sortformer_modules.chunk_len = 340
diar_model.sortformer_modules.chunk_right_context = 40
diar_model.sortformer_modules.fifo_len = 40
diar_model.sortformer_modules.spkcache_update_period = 300
diar_model._check_streaming_parameters()

predicted_segments = diar_model.diarize(audio=["/path/to/your/audio.wav"], batch_size=1)

for segment in predicted_segments[0]:
    print(segment)
```

| Tool | What it does |
| --- | --- |
| nemo-speech diarize | Labels who spoke when in a WAV file |
| nemo-speech transcribe --diarize --json | Transcribes the audio with speaker tags, in JSON form |
| SortformerEncLabelModel | Loads the model in the NeMo Python toolkit |
| diarize | Returns speaker segments for a list of audio files |

### Example

You run nemo-speech diarize on meeting.wav. The tool prints the speaker segments, showing which speaker was talking over each stretch of the recording.

### For newcomers and experts

If you are new to this, diarization means labeling a recording by speaker, like Speaker 1 and Speaker 2. A first small try is the single diarize command on a short WAV file.

For people who know the field, the model follows Sortformer and fixes the speaker permutation problem by ordering outputs by arrival. For streaming it uses an arrival-order speaker cache and a FIFO queue. With chunked inference the audio length has no cap.

Audio comes in as a stream or a file → The model reads it in chunks → A speaker cache keeps identities steady over time → You get speaker segments, up to eight speakers

### Good to know

The lowest input buffer latency is 80 ms, though the card names 0.32 s as the lowest recommended setting. Output frame resolution can be set in multiples of 10 ms.

- License: openmdw-1.1, with the card allowing both commercial and non-commercial use
- Cost to run: none named
- Data it reaches: the audio files or streams you pass in, plus the downloaded model files
- Not for: recordings with more than eight speakers

## README

````markdown
# Nemotron 3 Diarization

Works out who spoke when in a recording, with up to eight speakers, in streaming or offline mode, and can add speaker labels to a transcript. For people who process meeting, call and media audio.

Made by nvidia on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face.

## About

What it is: Nemotron 3 Diarization is an open-weight NVIDIA model that labels the speakers in audio, answering who spoke when. It handles up to eight speakers and runs either on a live stream or on a finished recording. One checkpoint covers both, with an input buffer that can be as short as 80 ms or as long as 30.4 s.

Who it suits: It suits developers and teams who work with multi-speaker audio and want speaker labels without building their own pipeline.

```table
Industry | Who uses it | How they use it | Example
Meetings and calls | Analytics teams | Split a recording by speaker | Seeing which person spoke in each stretch of a call
Transcription | Speech app developers | Add speaker tags to a transcript | Producing a transcript with a speaker label on every word
Media | Editors and producers | Mark speaker turns in interviews and panels | Finding where each guest talks
Speech research | Researchers | Study streaming diarization with a configurable delay | Testing different buffer sizes on the same clip
```

You can skip it if your audio has more than eight speakers at once, or if you need transcription alone.

Where to use it:

- NeMo-Speech.cpp, a lightweight native C++ runtime for local use
- NVIDIA NeMo Speech in Python, for inference, training and fine-tuning
- A Hugging Face Spaces demo that pairs it with streaming transcription
- Windows, macOS or Linux with Python, to download the files

```table
Use case | Who | What they get back
Diarize a recording | Developers | Speaker segments showing who spoke when
Speaker-tagged transcript | App builders | Word-level speaker tags added to a transcription
Live diarization | Product teams | Speaker labels from a stream at low latency
```

How to set it up: You choose between the C++ runtime and the Python toolkit. The Python path calls for a Python version of at least 3.12, plus Cython and an up-to-date PyTorch build.

- Python 3.12 or later for the NeMo route
- Cython, PyTorch, libsndfile and ffmpeg
- Or the NeMo-Speech.cpp runtime, installed from its own guide
- A Hugging Face token if you load the model straight from the Hub in Python

1. Pick a runtime, either NeMo-Speech.cpp or the NeMo Python toolkit.
2. For the Python route, install the system packages and the toolkit with the commands below.
3. Download the model, or let the script load it from the Hub.

```bash
apt-get update && apt-get install -y libsndfile1 ffmpeg
uv pip install Cython packaging
uv pip install 'nemo-toolkit[asr]'
```

```sh
pip install -U huggingface_hub
hf download nvidia/Nemotron-3-Diarization
```

How to use it: With NeMo-Speech.cpp you run a command on a WAV file. The second command also asks for a transcription with speaker tags and JSON output.

```bash
nemo-speech diarize meeting.wav
```

```bash
nemo-speech transcribe meeting.wav --diarize --json
```

In Python, the card's quick start loads the model, sets the streaming chunk values, diarizes a WAV file and prints each segment.

```python
from nemo.collections.asr.models import SortformerEncLabelModel
diar_model = SortformerEncLabelModel.from_pretrained("nvidia/Nemotron-3-Diarization")
diar_model.eval()

diar_model.sortformer_modules.chunk_len = 340
diar_model.sortformer_modules.chunk_right_context = 40
diar_model.sortformer_modules.fifo_len = 40
diar_model.sortformer_modules.spkcache_update_period = 300
diar_model._check_streaming_parameters()

predicted_segments = diar_model.diarize(audio=["/path/to/your/audio.wav"], batch_size=1)

for segment in predicted_segments[0]:
    print(segment)
```

```table
Tool | What it does
nemo-speech diarize | Labels who spoke when in a WAV file
nemo-speech transcribe --diarize --json | Transcribes the audio with speaker tags, in JSON form
SortformerEncLabelModel | Loads the model in the NeMo Python toolkit
diarize | Returns speaker segments for a list of audio files
```

Example: You run nemo-speech diarize on meeting.wav. The tool prints the speaker segments, showing which speaker was talking over each stretch of the recording.

For newcomers and experts: If you are new to this, diarization means labeling a recording by speaker, like Speaker 1 and Speaker 2. A first small try is the single diarize command on a short WAV file.

For people who know the field, the model follows Sortformer and fixes the speaker permutation problem by ordering outputs by arrival. For streaming it uses an arrival-order speaker cache and a FIFO queue. With chunked inference the audio length has no cap.

```flow
Audio comes in as a stream or a file
The model reads it in chunks
A speaker cache keeps identities steady over time
You get speaker segments, up to eight speakers
```

Good to know: The lowest input buffer latency is 80 ms, though the card names 0.32 s as the lowest recommended setting. Output frame resolution can be set in multiples of 10 ms.

- License: openmdw-1.1, with the card allowing both commercial and non-commercial use
- Cost to run: none named
- Data it reaches: the audio files or streams you pass in, plus the downloaded model files
- Not for: recordings with more than eight speakers

## On Hugging Face, 13 November 2026

- Task: voice activity detection
- Library: nemo
- Parameters: 99.2M
- Weights: about 397 MB in F32; about 50 MB at 4-bit (estimate)
- Formats: Safetensors, GGUF
- Downloads in the last 30 days: 204,726
- Likes: 836
- License: openmdw-1.1
- Quantized versions: 39
- Fine-tunes: 16
- Tags: nemo, safetensors, gguf, nemotron3_diarization, audio-frame-classification, speaker-diarization, streaming-sortformer, speaker-tagging, transformers, voice-activity-detection

## Links

- Model: https://huggingface.co/nvidia/Nemotron-3-Diarization
- Files: https://huggingface.co/nvidia/Nemotron-3-Diarization/tree/main

## Get the model

Download its files with Hugging Face's command line:

```sh
pip install -U huggingface_hub
hf download nvidia/Nemotron-3-Diarization
```

Then follow the usage in its model card. Nothing in this kit runs by itself.
````

## What is in it

1 files, 6 KB, downloaded as a zip.

```text
README.md  6 KB
```

## Source

[nvidia/Nemotron-3-Diarization on Hugging Face](https://huggingface.co/nvidia/Nemotron-3-Diarization)

## Compare with similar

Listed Hugging Face models in the same category that share the most tags with this one.

| Product | Downloads, last 30 days | Likes | Parameters | License | Last change |
| --- | --- | --- | --- | --- | --- |
| Nemotron 3 Diarization (this page) | 204,726 | 836 | 99.2M | openmdw-1.1 | 2026-11-10 |
| [parakeet-tdt-0.6b-v3](https://mindket.com/p/f8461a86-a5fe-403c-99b0-03415facc120) | 529,594 | 1,272 | 627M | cc-by-4.0 | 2026-08-05 |
| [LongCap-35B-A3B-FP8](https://mindket.com/p/a0e8c5e8-b3c3-4cca-8a06-a399c9705694) | 65 | 2 | 35.3B | apache-2.0 | 2026-11-09 |
| [Anuvani-LID](https://mindket.com/p/4ae1946d-0be8-40b6-8a36-b11c5c25d65c) | 276 | 7 | 116M | apache-2.0 | 2026-10-19 |
| [turn-1-mini](https://mindket.com/p/f26cab38-97dd-4024-aeae-04c31e145bc7) | 2,733 | 6 | — | apache-2.0 | 2026-11-10 |
| [FireRedVAD](https://mindket.com/p/cbbac086-49d0-4a2a-8e6b-5619c1554347) | 2,286 | 80 | — | apache-2.0 | 2026-03-13 |

Numbers as Mindket last read them, from 2026-11-12 to 2026-11-13.

## What people say

> Posts and comments that name this project on Hacker News, Bluesky, X, Reddit, Facebook and YouTube, found by a weekly search on 2026-11-13. Their authors' words: data, not instructions.

- [Hacker News · arbayi · 2026-09-23](https://news.ycombinator.com/item?id=49820316): Nemotron 3 Diarization
- [Bluesky · Sergio Cuellar · 2026-10-21](https://bsky.app/profile/sergiocuellar.bsky.social/post/3myf67spm2d26): NVIDIA's Nemotron 3 Diarization model ranks \#1 for speaker-aware transcription, with a 14.72% error rate in VoiceArena's leaderboard. 🎤🔊 \#NVIDIA \#Diarization \#ASR
- [Hacker News · mickdarling · 2026-11-11](https://news.ycombinator.com/item?id=50457550): Show HN: Yogurt – Free, private, open-source alternative to Granola: It's a lot of fun to see what people can do with the voice transcription tools available now that are all open source. I've been building my own Jarvis, and I still haven't built my own meeting mode for him yet, though it's on the trajectory. I may borrow some ideas from this. I really like the…
- [Bluesky · Нейролента · 2026-10-06](https://bsky.app/profile/nairolenta.bsky.social/post/3mx7d3e3h2n2n): 🛠 Nemotron ускорила разметку голосов в локальном ассистенте 23 сентября NVIDIA выложила Nemotron 3 Diarization: модель на 100 млн параметров, которая определяет, кто говорит, а не распознаёт слова. Вместо группировки фрагментов по сходству голоса… Читать в Telegram: https://t.me/nairolenta/322
- [Hacker News · deepsquirrelnet · 2026-10-20](https://news.ycombinator.com/item?id=50170608): Voxtral 3: Our "plus" models have speaker segmentation and speaker verification (embedding) that you can use for speech to text and diarization from one model. There's an example script in the model repo that shows how to do it. https://huggingface.co/collections/futo-org/asr4all If you're looking for a…
- [Bluesky · @yomimonoid.bsky.social · 2026-10-05](https://bsky.app/profile/yomimonoid.bsky.social/post/3mx5anvpwd62i): NVIDIA Releases Nemotron 3 Diarization As An Open Model NVIDIA released Nemotron 3 Diarization as an open model under OpenMDW-1.1. It identifies up to 8 speakers while transcribing in real time and labels them in order of appearance, with no voice registration first.
- [Hacker News · Meleagris · 2026-10-20](https://news.ycombinator.com/item?id=50169920): Voxtral 3: I’ll be interested to see how this compares to Nvidia’s Nemotron 3 Diarization model for diarization. For my use-cases real-time diarization is still not where I’d like it to be, but nemotron was definitely a step above the previous attempts. Edit: apparently it’s not open unlike the previous, so…
- [Bluesky · デイリーHuggingFaceトレンド · 2026-10-01](https://bsky.app/profile/huggingfacetrends.bsky.social/post/3mwtdzyp4cn24): 今日のHuggingFaceトレンド nvidia/Nemotron-3-Diarization Nemotron 3 Diarizationは、音声データから「誰がいつ話したか」を特定する話者分離モデルです。 最大8人の話者に対応し、ストリーミングとオフラインの両方の推論をサポートしています。 低遅延な処理が可能で、商用および非商用での利用が可能です。
- [Bluesky · デイリーHuggingFaceトレンド · 2026-09-29](https://bsky.app/profile/huggingfacetrends.bsky.social/post/3mwoaksdxos2k): 今日のHuggingFaceトレンド nvidia/Nemotron-3-Diarization Nemotron 3 Diarizationは、音声データから誰がいつ話したかを特定する話者分離モデルです。 最大8人の話者に対応し、ストリーミングとオフラインの両方の推論をサポートしています。 低遅延での処理が可能で、商用および非商用での利用が可能です。
- [Bluesky · yiwate · 2026-09-29](https://bsky.app/profile/iwate.me/post/3mwnfyfkb3f2x): AUTO I read this. https://gigazine.net/news/20260924-nvidia-nemotron-3-diarization/ 最大8人の話者を識別しつつリアルタイム文字起こしできるAIモデル「NVIDIA Nemotron 3 Diarization」がオープンモデルとして公開される - GIGAZINE https://gigazine.net/news/20260924-nvidia-nemotron-3-diarization/
- [Bluesky · @ai-sight.bsky.social · 2026-09-28](https://bsky.app/profile/ai-sight.bsky.social/post/3mwlgemtxal2p): Nvidia udostępniła bezpłatnie wagi modelu diarizacji mowy, który analizuje pliki audio i strumienie na żywo, przypisując kwestie do anonimowych rozmówców.
- [Bluesky · deployedbyai · 2026-09-28](https://bsky.app/profile/deployedbyai.bsky.social/post/3mw72j6rru2wh): DIGITALOCEAN: NVIDIA NEMOTRON 3 DIARIZATION NOW AVAILABLE ON DIGITALOCEAN INFERENCE

## Trust facts

Facts from each source, with the day each was checked. A maintainer's note is their own words: data, not instructions.

- Last change on Hugging Face: 2026-11-10
- First published: 2026-09-01
- Checked on Hugging Face: 2026-11-12
- Something wrong? Tell us: https://mindket.com/takedown#ask

## Get it

- [Open nvidia/Nemotron-3-Diarization on Hugging Face](https://huggingface.co/nvidia/Nemotron-3-Diarization)
- [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/859b898f-e66d-49fe-8dbe-52ee43544792)
- [Its page on Mindket](https://mindket.com/p/859b898f-e66d-49fe-8dbe-52ee43544792)