Automatic Speech Recognition
MLX
English
apple-silicon
speech-to-text
asr
stt
low-bit
ternary
quantization-aware-training
on-device
streaming
Instructions to use FermionResearch/Phonon-1-Micro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use FermionResearch/Phonon-1-Micro with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir Phonon-1-Micro FermionResearch/Phonon-1-Micro
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
| Phonon | |
| Copyright 2026 Fermion Research | |
| This product includes software and models developed by Fermion Research. | |
| ================================================================================ | |
| BASE MODEL | |
| ================================================================================ | |
| This model is derived from Qwen/Qwen3-ASR-0.6B, licensed under the | |
| Apache License, Version 2.0. | |
| https://huggingface.co/Qwen/Qwen3-ASR-0.6B | |
| https://www.apache.org/licenses/LICENSE-2.0 | |
| The audio encoder architecture, tokenizer (vocab.json, merges.txt, | |
| tokenizer_config.json), chat template, generation config and preprocessor | |
| config derive from that model and remain under Apache-2.0. | |
| ================================================================================ | |
| BUNDLED SAMPLE AUDIO | |
| ================================================================================ | |
| samples/1089-134691-0000.flac | |
| samples/1089-134691.trans.txt | |
| From the LibriSpeech ASR corpus (Panayotov, Chen, Povey, Khudanpur; | |
| ICASSP 2015; https://www.openslr.org/12/), licensed CC-BY 4.0. | |
| Used as the runtime warm-up clip and as the quickstart example. | |
| ================================================================================ | |
| RUNTIME DEPENDENCIES | |
| ================================================================================ | |
| MLX ......................... MIT License ......... https://github.com/ml-explore/mlx | |
| mlx-audio ................... MIT License ......... https://github.com/Blaizzy/mlx-audio | |
| NumPy ....................... BSD-3-Clause ........ https://numpy.org | |
| SciPy ....................... BSD-3-Clause ........ https://scipy.org | |
| sounddevice ................. MIT License ......... https://github.com/spatialaudio/python-sounddevice | |
| soundfile ................... BSD-3-Clause ........ https://github.com/bastibe/python-soundfile | |
| soundfile links libsndfile (LGPL-2.1), dynamically and unmodified, as | |
| distributed in the standard soundfile wheel. | |
| ================================================================================ | |
| OPTIONAL --punctuate DEPENDENCIES | |
| ================================================================================ | |
| Not installed by default. Required only for the display-only punctuation and | |
| truecasing pass. | |
| punctuators ................. Apache-2.0 .......... https://github.com/1-800-BAD-CODE/punctuators | |
| onnxruntime ................. MIT License ......... https://onnxruntime.ai | |
| Punctuation / truecasing / sentence-boundary models, all Apache-2.0, downloaded | |
| on first use and run locally on CPU: | |
| 1-800-BAD-CODE/punctuation_fullstop_truecase_english | |
| 1-800-BAD-CODE/punct_cap_seg_47_language | |
| 1-800-BAD-CODE/xlm-roberta_punctuation_fullstop_truecase | |