FunASR README

The project's README from github.com/modelscope/FunASR. As one markdown file: README.md.

(简体中文|English|日本語|한국어)

FunASR

Industrial speech recognition toolkit for offline, streaming, and edge deployment.
ASR · VAD · punctuation · speaker pipelines · emotion and audio-event models · OpenAI-compatible serving

PyPI Stars Downloads Docs MCP Toplist

modelscope%2FFunASR | Trendshift

Quick Start · Model selection · Models · Deployment matrix · Deployment hub · Docs · Benchmark · Contribute

Quick Start

Native Transformers

For Fun-ASR-Nano transcription with the Hugging Face API, start with the Transformers 5.17.0 CPU quickstart. No FunASR toolkit or remote Python code is needed.

Space · Notebook · Python / batch examples

FunASR toolkit and pipelines

Open In Colab

No local setup? Open the Colab quickstart to transcribe a public sample or upload your own audio in a browser.

Found FunASR useful? Star the project so more builders can find it.

# CPU-only installs can use the default PyPI wheels.
pip install torch torchaudio
pip install funasr

For GPU quickstarts, install the PyTorch and torchaudio wheels that match your NVIDIA driver from pytorch.org before installing FunASR. After installation, confirm the GPU is visible:

python - <<'PY'
import torch
print(torch.cuda.is_available())
PY

Only use device="cuda" when this prints True; otherwise use device="cpu" or reinstall PyTorch with the correct CUDA wheel.

FunASR toolkit GPU example: Fun-ASR-Nano (Chinese, English, Japanese, and Chinese dialect groups and regional accents; the separate native Transformers CPU path is linked above):

from funasr import AutoModel

model = AutoModel(model="FunAudioLLM/Fun-ASR-Nano-2512", device="cuda")
result = model.generate(input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav")
print(result[0]["text"])

For the separate 31-language checkpoint, use Fun-ASR-MLT-Nano-2512. Language coverage is checkpoint-specific, so Nano and MLT-Nano should be treated as distinct model choices.

For a CPU-first example with five-language ASR plus emotion and audio-event tags, use SenseVoiceSmall. The pipeline below combines it with FSMN-VAD and CAM++ for speaker-aware VAD segments; these are not native speaker outputs of the SenseVoiceSmall checkpoint. See the SenseVoice paper, Hugging Face checkpoint, and GGUF edge checkpoint.

from funasr import AutoModel
from funasr.utils.postprocess_utils import rich_transcription_postprocess

model = AutoModel(model="iic/SenseVoiceSmall", vad_model="fsmn-vad", spk_model="cam++", device="cpu")
result = model.generate(
    input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav",
    batch_size_s=300,
)

# The AutoModel pipeline returns VAD segments with speaker ids and timestamps:
for seg in result[0]["sentence_info"]:
    print(f"[{seg['start']/1000:.1f}s] Speaker {seg['spk']}: {rich_transcription_postprocess(seg['sentence'])}")

This prints each returned segment's start time in seconds, anonymous speaker index, and text with SenseVoice tags removed. Text and segment boundaries depend on the audio and checkpoint; no fixed transcript is asserted here.

CAM++ extracts spk_embedding vectors. AutoModel clusters those embeddings and assigns speaker indices to VAD segments. Indices are local to a recording, not known-person identities. See the SDK contract for the component and result boundaries. Change to device="cuda" only after verifying a compatible GPU environment as described above.

Scale & deploy the flagship

At scale, accelerate Fun-ASR-Nano with vLLM (batch processing):

from funasr.auto.auto_model_vllm import AutoModelVLLM

model = AutoModelVLLM(model="FunAudioLLM/Fun-ASR-Nano-2512", tensor_parallel_size=1)
results = model.generate(["audio1.wav", "audio2.wav"], language="auto")

Deploy as API server: Local SenseVoice CPU recipe · Nano GPU serving and pinned vLLM setup

Use with AI agents: MCP Server for Claude/Cursor · OpenAI API for LangChain/Dify/AutoGen

Use with voice agents: OpenClaw realtime plugin for self-hosted Talk and Voice Call transcription

Why FunASR?

FunASR is a toolkit: choose the task, checkpoint, and runtime separately. Support in one model or adapter does not imply support in every serving backend.

Task Checkpoint or pipeline Runtime entrypoint Important limitation
File transcription with emotion/event tags SenseVoiceSmall Python AutoModel, CPU or GPU Five-language checkpoint; tags do not identify speakers.
LLM-based file transcription Fun-ASR-Nano AutoModel; split-engine AutoModelVLLM for the documented GPU path Base Nano covers zh/en/ja and Chinese dialects/accents; timestamp support depends on checkpoint and path.
Broader multilingual transcription Fun-ASR-MLT-Nano Python AutoModel Separate 31-language checkpoint; do not transfer its coverage to base Nano.
Chunked live transcription Paraformer-zh-streaming Streaming SDK or runtime WebSocket service Use the streaming checkpoint and per-session cache, not an offline checkpoint.
Speaker-aware file transcription SenseVoiceSmall + FSMN-VAD + CAM++ AutoModel with VAD and embedding clustering Anonymous indices within a recording, not enrolled-speaker identification.
Joint text, timestamps, and speakers MOSS-Transcribe-Diarize, third-party OpenMOSS FunASR adapter or upstream backend in the MOSS guide Offline, recording-local anonymous labels; no external VAD/speaker pipeline for its unified path.
Native CPU/edge transcription Fun-ASR-Nano or SenseVoiceSmall GGUF llama.cpp runtime Requires matching converted weights; GGUF is not a Python AutoModel checkpoint.

See the Model Zoo and deployment matrix for checkpoint, interface, and licensing boundaries. Benchmark on your own audio and hardware before choosing a runtime.

Trying FunASR for the first time? Use the Colab quickstart before setting up a local environment. Choosing a first model? Start with the model selection guide. Planning a switch from Whisper or a cloud ASR provider? Use the migration guide and benchmark example to test representative audio, map features, and roll out safely.

Installation

pip install funasr
From source / Requirements 
git clone https://github.com/modelscope/FunASR.git && cd FunASR
pip install -e ./

Requirements: Python ≥ 3.8. Install PyTorch + torchaudio first (pytorch.org), then pip install funasr.

Model Zoo

This list includes third-party models. OpenMOSS publishes MOSS-Transcribe-Diarize; FunASR provides an adapter, not ownership of its weights. Its unified path is offline, with anonymous labels scoped to each recording, not realtime or known-person identification. Model licenses are separate from the toolkit's MIT license.

Model Task Languages Params Links
Fun-ASR-Nano ASR zh/en/ja + Chinese dialects and accents 800M ⭐ HF / Transformers · HF / FunASR GGUF
Fun-ASR-MLT-Nano ASR 31 languages 800M ⭐ 🤗
SenseVoiceSmall ASR + emotion + events zh/en/ja/ko/yue 234M ⭐ 🤗 GGUF paper
MOSS-Transcribe-Diarize Third-party OpenMOSS: offline ASR + timestamps + anonymous speakers See official card See official card 🤗 guide
Paraformer-zh ASR + timestamps zh/en 220M ⭐ 🤗
Paraformer-zh-streaming Streaming ASR zh/en 220M ⭐ 🤗
Qwen3-ASR ASR, 52 languages multilingual 1.7B usage
GLM-ASR-Nano ASR, 17 languages multilingual 1.5B usage
Whisper-large-v3 ASR + translation multilingual 1550M usage
Whisper-large-v3-turbo ASR + translation multilingual 809M usage
ct-punc Punctuation zh/en 290M ⭐ 🤗
fsmn-vad VAD zh/en 0.4M ⭐ 🤗
cam++ Speaker embeddings (pipeline component) — 7.2M ⭐ 🤗
emotion2vec+large Emotion recognition — 300M ⭐ 🤗

Usage

Python tutorial · SDK parameters and outputs · Training · Model registration

from funasr import AutoModel

# Chinese production (VAD + ASR + punctuation + speaker)
model = AutoModel(model="paraformer-zh", vad_model="fsmn-vad", punc_model="ct-punc", spk_model="cam++", device="cuda")
result = model.generate(input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav", hotword="关键词 20")

# Optional Silero VAD (install first: python -m pip install "funasr[silero]")
model = AutoModel(
    model="paraformer-zh", vad_model="silero-vad", device="cuda",
    vad_kwargs={"silero_threshold": 0.5, "silero_min_silence_duration_ms": 100},
)
result = model.generate(input="audio.wav")

# Streaming real-time (feed audio chunk by chunk)
import soundfile as sf
model = AutoModel(model="paraformer-zh-streaming", device="cuda")
audio, sr = sf.read("speech.wav", dtype="float32")   # 16 kHz mono
chunk_size = [0, 10, 5]                               # 600 ms chunks
chunk_stride = chunk_size[1] * 960
cache = {}
n_chunks = (len(audio) - 1) // chunk_stride + 1
for i in range(n_chunks):
    chunk = audio[i * chunk_stride : (i + 1) * chunk_stride]
    res = model.generate(input=chunk, cache=cache, is_final=(i == n_chunks - 1),
                         chunk_size=chunk_size, encoder_chunk_look_back=4, decoder_chunk_look_back=1)
    if res[0]["text"]:
        print(res[0]["text"], end="", flush=True)

# Emotion recognition
model = AutoModel(model="emotion2vec_plus_large", device="cuda")
result = model.generate(input="audio.wav", granularity="utterance")

CLI (Agent-Friendly)

# Transcribe audio (simplest)
funasr audio.wav

# JSON output (for AI agents)
funasr audio.wav --output-format json

# SRT subtitles
funasr audio.wav --output-format srt --output-dir ./subs

# Speaker diarization + timestamps
funasr audio.wav --spk --timestamps -f json

# Choose model and language
funasr audio.wav --model paraformer --language zh

# Batch transcribe
funasr *.wav --output-format srt --output-dir ./output

Available models: sensevoice (default), paraformer, paraformer-en, fun-asr-nano

Deploy

Start a local SenseVoice CPU service from a fresh directory in a POSIX shell with Python 3.11. This installs the PyPI release into a separate environment, not this source checkout. Keep the unauthenticated service on loopback; use the security guide before exposing it to other clients.

python3.11 -m venv .venv-funasr-http
. .venv-funasr-http/bin/activate
python -m pip install torch torchaudio
python -m pip install funasr fastapi uvicorn python-multipart
python -m pip check
funasr-server --host 127.0.0.1 --port 8000 --model sensevoice --device cpu

Wait for model download and server startup. In a second terminal, use the same directory and curl 7.76+ to download a public Chinese audio sample and transcribe it. The request uses the preloaded model; no fixed transcript or speaker labels are promised.

curl --fail --location https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/BAC009S0764W0121.wav -o sample.wav && \
curl --fail-with-body http://127.0.0.1:8000/v1/audio/transcriptions \
  -F [email protected] \
  -F model=sensevoice \
  -F response_format=verbose_json

For offline joint ASR and anonymous speaker labels (moss-transcribe-diarize), prepare the separate environment in the MOSS service, Docker, Kubernetes, vLLM, SGLang, LocalAI, and FunClip guide →. It is an alternative service, not another command in the CPU environment. Stop the CPU service before reusing port 8000. For Nano GPU serving, follow the pinned split-engine guide and inspect the actual backend logs; selecting a model does not by itself prove that vLLM was loaded.

# Docker streaming service
docker pull registry.cn-hangzhou.aliyuncs.com/funasr_repo/funasr:funasr-runtime-sdk-online-cpu-0.1.12

CPU / Edge — llama.cpp / GGUF (no GPU, no Python)

Run SenseVoice / Paraformer / Fun-ASR-Nano as a single self-contained binary on CPU and edge devices — this is to FunASR what whisper.cpp is to Whisper, but with ~3× lower CER than whisper.cpp on Chinese. Built-in FSMN-VAD, no Python at runtime.

# Linux / macOS: run from the extracted release directory
bash download-funasr-model.sh sensevoice ./gguf        # or: paraformer | nano
./llama-funasr-sensevoice -m ./gguf/sensevoice-small-q8.gguf --vad ./gguf/fsmn-vad.gguf -a audio.wav
# → 欢迎大家来体验达摩院推出的语音识别模型

# Windows PowerShell: run from the extracted archive root (with the `hf` CLI installed)
hf download FunAudioLLM/SenseVoiceSmall-GGUF sensevoice-small-q8.gguf --local-dir .\gguf
hf download FunAudioLLM/fsmn-vad-GGUF fsmn-vad.gguf --local-dir .\gguf
.\llama-funasr-sensevoice.exe -m .\gguf\sensevoice-small-q8.gguf --vad .\gguf\fsmn-vad.gguf -a audio.wav
# Use the windows-x64-vulkan package with a current AMD, Intel, or NVIDIA Vulkan driver:
.\llama-funasr-sensevoice.exe -m .\gguf\sensevoice-small-q8.gguf --vad .\gguf\fsmn-vad.gguf -a audio.wav --backend vulkan
# Use the windows-x64-cuda package on RTX 30-class GPUs:
.\llama-funasr-sensevoice.exe -m .\gguf\sensevoice-small-q8.gguf --vad .\gguf\fsmn-vad.gguf -a audio.wav --backend cuda

Use funasr-llamacpp-linux-x64-vulkan.tar.gz on Linux GPU systems with a working Vulkan driver/ICD:

./llama-funasr-sensevoice -m ./gguf/sensevoice-small-q8.gguf --vad ./gguf/fsmn-vad.gguf -a audio.wav --backend vulkan

The Windows Vulkan ZIP uses the system Vulkan loader supplied by the GPU driver; installing the Vulkan SDK is only necessary when building from source. Both Vulkan packages currently accelerate SenseVoiceSmall.

Tagged releases provide two Windows CUDA packages. The standard windows-x64-cuda ZIP targets CUDA architecture 86, while windows-x64-cuda-blackwell targets architecture 120 (sm_120) for RTX 50 / Blackwell GPUs. Both ZIPs bundle the required cuBLAS DLLs and use the static MSVC runtime, so users need a compatible NVIDIA driver but not a separate CUDA Toolkit installation. CI verifies the architecture and package boundary; it does not prove inference on physical Blackwell hardware.

Prebuilt binaries: Releases · v0.2.6 · Linux Vulkan tarball · Windows Vulkan zip · Windows CUDA zip · Windows Blackwell CUDA zip · Download & quickstart: funasr.com/deploy/llama-cpp · GGUF models: Hugging Face · Docs & benchmarks: runtime/llama.cpp/

OpenAI API example → · [Gradio demo →](./examples/openai_api

FunASR (modelscope/FunASR)

Searching Database…

20.7K ·GitHub
FunASR
What is this?FunASR is an open-source industrial speech recognition toolkit covering ASR, VAD, punctuation, speaker diarization, and emotion/audio-event models, deployable offline, in streaming mode, on edge devices, or behind an OpenAI-compatible API.
Is it a Python library, a CLI, or a server?All three in one toolkit: a pip-installable Python library (AutoModel and pipelines), an agent-friendly CLI, and serving options — an OpenAI-compatible API server, an MCP server, and standalone llama.cpp binaries.
How do I get it running?Fastest is the Colab quickstart — transcribe audio in a browser with zero local setup. Locally, install Python 3.8+, add PyTorch and torchaudio from pytorch.org, then pip install funasr.
Do I need PyTorch, and is a GPU required?PyTorch and torchaudio are required for the Python toolkit; a GPU is optional. SenseVoiceSmall runs CPU-first, and you should only pass device='cuda' after the GPU check prints True — otherwise stay on CPU or reinstall PyTorch with the matching CUDA wheel.
What pitfalls should I know about when picking a model?Language coverage is checkpoint-specific: Fun-ASR-Nano covers zh/en/ja plus Chinese dialects, while Fun-ASR-MLT-Nano is a separate 31-language checkpoint — one's coverage does not transfer to the other. Speaker labels are anonymous per-recording indices, not known-person identification.
Does it work on Windows, or is it Linux-only?The Python package installs wherever Python 3.8+ runs, and there are prebuilt Windows binaries: Vulkan and CUDA ZIPs, including a separate sm120 build for RTX 50/Blackwell GPUs. They bundle cuBLAS DLLs, so an NVIDIA driver is needed but not the CUDA Toolkit.
Can it run fully offline?Yes. Offline file transcription runs locally with any checkpoint, and the GGUF/llama.cpp runtime ships as a self-contained binary for CPU and edge devices with built-in FSMN-VAD — no Python at runtime.
Where does my audio go when I use the API server?The quickstart service is unauthenticated and meant to stay on loopback; the README says to consult the security guide before exposing it to other clients, so audio stays on your machine unless you deliberately open the server up.
Is everything under the MIT license?The toolkit is MIT, but each model in the Model Zoo carries its own license — check the deployment matrix for checkpoint-level licensing boundaries. Third-party models like MOSS-Transcribe-Diarize are provided via an adapter, not FunASR's own weights.
Can it transcribe live audio in real time?Yes — chunked live transcription runs on Paraformer-zh-streaming through the Streaming SDK or a runtime WebSocket service. You must use the streaming checkpoint with a per-session cache; an offline checkpoint will not work for this.
How does it compare to Whisper?The README pitches the llama.cpp/GGUF runtime as FunASR's answer to whisper.cpp, claiming about 3x lower CER than whisper.cpp on Chinese. The Model Zoo also runs Whisper-large-v3 and its turbo checkpoint, and a migration guide helps you switch from Whisper or a cloud ASR provider.
Which models are in the Model Zoo?ASR checkpoints from 234M SenseVoiceSmall (zh/en/ja/ko/yue plus emotion and audio-event tags) up to 1.7B Qwen3-ASR covering 52 languages, plus Whisper-large-v3, Paraformer-zh, ct-punc for punctuation, fsmn-vad, cam++ speaker embeddings, and emotion2vec+large for emotion recognition.
How do I hook it up to AI agents?Deploy it as an OpenAI-compatible API for LangChain, Dify, and AutoGen, connect it via the MCP Server for Claude and Cursor, batch-process with vLLM, or use the OpenClaw realtime plugin for self-hosted talk and voice-call transcription.
What are similar projects?WhisperLiveKit, VibeVoice, Qwen3-ASR, DeepSpeech, hayamimi, speechbrain, Whisper, TalkLikeYou-Streaming-RealTime-Demo
Which topics is it in?emotion-recognition, voice-activity-detection, speaker-diarization, asr, llama-cpp, transcription, gguf, speech-to-text, speech-recognition, vllm, openai-compatible, multilingual
Creator
modelscope
Size
124.9 MB
Stars
20.7K
Updated
CommitOct 17Releasev1.4.16Sep 18
Age
3y
Commits · 90 Days