Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,27 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
language:
|
| 4 |
+
- ja
|
| 5 |
+
library_name: mlx
|
| 6 |
+
tags:
|
| 7 |
+
- whisper
|
| 8 |
---
|
| 9 |
+
|
| 10 |
+
# kotoba-whisper-v1.0-mlx
|
| 11 |
+
|
| 12 |
+
This repository contains a converted `mlx-whisper` model of [kotoba-whisper-v1.0](https://huggingface.co/kotoba-tech/kotoba-whisper-v1.0) which is suitable for running with Apple Silicon.
|
| 13 |
+
As `kotoba-whisper-v1.0` is derived from `distil-large-v3`, this model is significantly faster than [mlx-community/whisper-large-v3-mlx](https://huggingface.co/mlx-community/whisper-large-v3-mlx) without losing much accuracy for Japanese transcription.
|
| 14 |
+
|
| 15 |
+
## Usage
|
| 16 |
+
|
| 17 |
+
```sh
|
| 18 |
+
pip install mlx-whisper
|
| 19 |
+
```
|
| 20 |
+
|
| 21 |
+
```py
|
| 22 |
+
mlx_whisper.transcribe(speech_file, path_or_hf_repo="kaiinui/kotoba-whisper-v1.0-mlx")
|
| 23 |
+
```
|
| 24 |
+
|
| 25 |
+
## Related Links
|
| 26 |
+
* [kotoba-whisper-v1.0](https://huggingface.co/kotoba-tech/kotoba-whisper-v1.0) (The original model)
|
| 27 |
+
* [mlx-whisper](https://github.com/ml-explore/mlx-examples/tree/main/whisper)
|