---
title: Phonon-2.5
canonical_url: "https://www.modelscope.cn/models/FermionResearch/Phonon-2.5"
md_url: "https://www.modelscope.cn/models/FermionResearch/Phonon-2.5.md"
repository: FermionResearch/Phonon-2.5
last_updated: 2026-10-29
license: cc-by-4.0
model_type:
  - parakeet_tdt_five_value
base_model:
  - nvidia/parakeet-tdt-0.6b-v3
base_model_relation: quantized
library_name:
  - mlx
  - pytorch
frameworks:
  - pytorch
language:
  - en
  - es
  - fr
  - de
  - it
  - pt
  - nl
  - pl
  - ru
  - uk
  - ro
  - cs
  - sk
  - bg
  - hr
  - sv
  - da
  - fi
  - et
  - lv
  - lt
  - hu
  - sl
  - mt
  - el
downloads: 0
stars: 0
tags:
  - mlx
  - apple-silicon
  - speech-to-text
  - asr
  - stt
  - multilingual
  - low-bit
  - on-device
---

# Phonon-2.5

> Phonon-2.5 - FermionResearch 在 ModelScope 开源的模型。Phonon-2.5 is an open speech recognition model for 25 languages in a 156.2 MB download. On the FLEURS test sets it averages 11.71 % word error across the 25 languages, against 12.18 % for Parakeet Redux, the…

FermionResearch/Phonon-2.5 是 ModelScope 魔搭社区上的机器学习模型，采用 cc-by-4.0 许可，基于 nvidia/parakeet-tdt-0.6b-v3 构建。

- **Repository**: FermionResearch/Phonon-2.5
- **License**: cc-by-4.0
- **Base model**: nvidia/parakeet-tdt-0.6b-v3
- **Tags**: mlx, apple-silicon, speech-to-text, asr, stt, multilingual, low-bit, on-device
- **Downloads**: 0
- **Stars**: 0
- **Last updated**: 2026-10-29

Source: https://www.modelscope.cn/models/FermionResearch/Phonon-2.5

---

![Phonon-2.5](https://huggingface.co/FermionResearch/Phonon-2.5/resolve/main/phonon-2.5-card.jpg)

# Phonon-2.5

Phonon-2.5 is an open speech recognition model for 25 languages in a 156.2 MB download. On the FLEURS test sets it
averages 11.71 % word error across the 25 languages, against 12.18 % for Parakeet Redux, the other low-bit model of the same size,
and 13.26 % for Parakeet TDT 0.6B v3, the 2.5 GB full-precision model it derives from. It is ahead of Redux in 14 of the 25
languages. The same file runs on Apple silicon and on Linux and Windows CPUs, with no GPU needed.

For English alone, see [Phonon-2.5 English](https://huggingface.co/FermionResearch/Phonon-2.5-English), the same size, English only.

## Benchmarks

Word error rate, %, lower is better; the best value in each row is in bold. All three models were scored on the full FLEURS test set of
each language with the same harness and the same text normalisation.

| Language | Phonon-2.5 (156.2 MB) | Parakeet Redux (178 MB) | Parakeet TDT 0.6B v3 (2,508 MB) |
|---|--:|--:|--:|
| English | 6.58 | 7.32 | **6.53** |
| Spanish | 4.80 | 4.99 | **4.69** |
| French | 8.12 | 9.03 | **6.61** |
| German | 6.67 | 7.57 | **5.96** |
| Italian | 5.21 | 5.12 | **4.75** |
| Portuguese | 6.53 | 6.44 | **6.27** |
| Dutch | 9.21 | 9.45 | **7.93** |
| Polish | 10.20 | 10.19 | **8.55** |
| Russian | 8.67 | 9.01 | **7.21** |
| Ukrainian | 7.89 | 8.05 | **7.27** |
| Romanian | **11.88** | 11.89 | 13.13 |
| Czech | **10.40** | 11.41 | 11.93 |
| Slovak | 8.89 | **8.17** | 9.72 |
| Bulgarian | 13.34 | **13.02** | 13.48 |
| Croatian | 12.24 | **11.43** | 13.21 |
| Swedish | **14.53** | 14.70 | 15.44 |
| Danish | 18.20 | **17.91** | 19.05 |
| Finnish | 12.02 | **11.84** | 13.08 |
| Estonian | 11.91 | **10.29** | 14.34 |
| Latvian | 14.92 | **14.76** | 23.66 |
| Lithuanian | **17.94** | 18.05 | 21.68 |
| Hungarian | 15.45 | 15.81 | **15.38** |
| Slovenian | **18.53** | 18.88 | 24.70 |
| Maltese | 17.91 | **16.02** | 20.86 |
| Greek | **20.83** | 33.25 | 36.11 |
| **Average of 25** | **11.71** | 12.18 | 13.26 |

On the Open ASR Leaderboard's seven English sets (LibriSpeech clean and other, AMI, Earnings-22, GigaSpeech, SPGISpeech,
VoxPopuli, scored with its code on the full test sets) Phonon-2.5 averages 5.06 % word error: 1.60, 3.41, 9.12, 6.72, 8.04, 3.31,
3.23. Phonon-2 averages 5.21 % on the same sets; Phonon-2.5 English averages 4.82 %.

## Run it

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

```bash
pip install fermion-research
fermion transcribe phonon-2.5 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:

```bash
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 --port 8010` starts an OpenAI-compatible
`/v1/audio/transcriptions` endpoint. The command line and
the server are documented at [fermionresearch.com/docs/speech](https://www.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](https://pypi.org/project/fermion-research/)
and this repository's code are Apache 2.0.