中文Sign up

FluidInference/qwen3-asr-0.6b-coreml

🤗 Hugging Face sourceautomatic-speech-recognitionapache-2.0600M activated5.0 GBother✓ 32 checksumsupdated today
Submit in one command

Run it next to your model folder. It makes the torrent, checks your files against Hugging Face, and submits it. You just start seeding and paste your key from your account. It only reads your files and never changes them. Read the script first if you like.

curl -fsSL https://pirateface.co/package.sh | bash -s -- --repo FluidInference/qwen3-asr-0.6b-coreml ./model-folder
Needs a seeder →

license: apache-2.0 language:

  • en
  • zh
  • ja
  • ko
  • vi
  • th
  • id
  • ms
  • hi
  • ar
  • tr
  • ru
  • de
  • fr
  • es
  • multilingual tags:
  • speech-recognition
  • asr
  • coreml
  • apple
  • ios
  • macos
  • qwen
  • audio library_name: coreml pipeline_tag: automatic-speech-recognition base_model: Qwen/Qwen3-ASR-0.6B

Qwen3-ASR 0.6B CoreML

Core ML conversion of Qwen/Qwen3-ASR-0.6B for on-device speech recognition on Apple platforms (iOS/macOS).

Model Variants

Variant Size Description
f32/ ~2.5 GB Full precision (Float32) - highest accuracy
int8/ ~0.7 GB Quantized (Int8) - smaller, faster

Features

  • 30+ languages including English, Chinese, Japanese, Korean, and more
  • On-device inference - no internet required
  • Autoregressive decoder with KV-cache support
  • Processes audio in 1-second chunks (100 mel frames)

Benchmarks (M4 Pro)

Dataset WER CER RTFx
LibriSpeech test-clean (2620 files) 4.4% 1.9% 2.8x
AISHELL-1 test (100 files) 4.6% 3.7% 4.5x

Official PyTorch model: 2.11% WER on LibriSpeech test-clean

Usage with FluidAudio

import FluidAudio

let manager = Qwen3AsrManager()
try await manager.loadModels()

let samples = try AudioConverter().resampleAudioFile(path: "audio.wav")
let transcript = try await manager.transcribe(
    audioSamples: samples,
    language: "en",
    maxNewTokens: 512
)
print(transcript)

Model Architecture

  • Encoder: Audio encoder (Whisper-style mel spectrogram input)
  • Decoder: 28-layer transformer decoder with 1024 hidden size
  • Tokenizer: Qwen tokenizer with special ASR tokens

License

Apache 2.0 - Same as the original Qwen3-ASR model.

Credits

Citation

@article{qwen3asr, title={Qwen3-ASR Technical Report}, author={Qwen Team}, journal={arXiv preprint arXiv:2601.21337}, year={2025} }

For the HuggingFace metadata UI, fill in:

  • License: Apache 2.0
  • Base model: Qwen/Qwen3-ASR-0.6B
  • Pipeline: automatic-speech-recognition
  • Library: coreml
  • Languages: en, zh, ja, ko, + others