Phonon-2.5-English

license: cc-by-4.0 thumbnail: https://huggingface.co/FermionResearch/Phonon-2.5-English/resolve/main/phonon-2.5-english-card.jpg base_model: nvidia/parakeet-tdt-0.6b-v3 base_model_relation: quantized language:

en library_name: mlx pipeline_tag: automatic-speech-recognition tags:
mlx
apple-silicon
speech-to-text
asr
stt
low-bit
on-device metrics:
wer

Phonon-2.5 English

Phonon-2.5 English is an open speech recognition model for English in a 157.6 MB download. Across the Open ASR Leaderboard's seven English sets it averages 4.82 % word error, against 4.94 % for Underdog Bark 3.0, the other low-bit model of this size, and 5.21 % for Phonon-2. It is better than Phonon-2 on all seven sets. The same file runs on Apple silicon and on Linux and Windows CPUs, with no GPU needed.

Why a separate English model? Some workloads want English text and nothing else: dictation, captions, compliance and search pipelines built around English. Phonon-2.5 English is for those, and it is our most accurate English model.

For 25 languages, see Phonon-2.5, the same size.

Benchmarks

Word error rate, %, lower is better; the best value in each column is in bold.

Model	Download	LS clean	LS other	AMI	Earnings-22	GigaSpeech	SPGISpeech	VoxPopuli	Average
Phonon-2.5 English	157.6 MB	1.64	3.56	8.76	6.22	7.99	3.31	2.28	4.82
Phonon-2.5 (25 languages)	156.2 MB	1.60	3.41	9.12	6.72	8.04	3.31	3.23	5.06
Phonon-2	158 MB	1.72	3.92	9.37	6.96	8.35	3.70	2.46	5.21
Underdog Bark 3.0	163 MB	1.89	4.08	7.82	7.05	7.95	3.72	2.05	4.94
Parakeet TDT 0.6B v3†	2,508 MB	1.52	3.13	9.42	5.85	7.99	3.63	3.19	4.96
Parakeet Redux	178 MB	1.94	4.35	9.16	7.90	8.62	4.01	3.87	5.69

† Open ASR Leaderboard's published row; the other rows were scored with its code on the full test sets, with Earnings-22 as the leaderboard's chunked set. The Phonon-2 row is the one published with its release.

Run it

Python 3.10 or newer (macOS's built-in python3 is 3.9).

pip install fermion-research
fermion transcribe phonon-2.5-en recording.wav

On Apple silicon, pip install "fermion-research[mlx]" runs the model on the GPU through MLX. On Linux (x86-64 and Arm) and Windows the same package runs the CPU engine; on Linux, install torch from its CPU wheel index first to skip the GPU build:

pip install --no-deps torch --index-url https://download.pytorch.org/whl/cpu
pip install fermion-research

--json adds a start and end time for every word. fermion serve phonon-2.5-en --port 8010 starts an OpenAI-compatible /v1/audio/transcriptions endpoint. The command line and the server are documented at fermionresearch.com/docs/speech.

Notes

Based on parakeet-tdt-0.6b-v3 by NVIDIA; the tokenizer and output conventions (punctuation, casing, numerals) are the original's. Licence CC-BY-4.0, same as the original; NOTICE lists the changes. The command line and this repository's code are Apache 2.0.