Downloads · 30 days
23
14% of all-time downloads
doggy8088/whisper-large-v2-cantonese-mlx
whisper-large-v2-cantonese-mlx is a automatic speech recognition model from doggy8088. Use it when you need speech turned into text. It is set up for mlx. The card lists the license as apache-2.0.
這是 Scrya/whisper-large-v2-cantonese 的 MLX 轉換版本。 原始模型是以 openai/whisper-large-v2 為基礎,並在 Common Voice 11.0 粵語資料集上微調的 Cantonese ASR 模型。
Downloads · 30 days
23
14% of all-time downloads
All-time downloads
170
Public
Repo size
3.1 GB
Likes
0
Public
Click a slice to open those files.
.safetensors3.1 GB · 100%
From the Hugging Face model README
這是 Scrya/whisper-large-v2-cantonese 的 MLX 轉換版本。 原始模型是以 openai/whisper-large-v2 為基礎,並在 Common Voice 11.0 粵語資料集上微調的 Cantonese ASR 模型。
根據原始模型卡,它在 mozilla-foundation/common_voice_11_0 的 yue 測試集 CER 約為 6.21。
先安裝:
pip install -U mlx-whisper
CLI:
mlx_whisper audio.wav --model doggy8088/whisper-large-v2-cantonese-mlx --language zh
Python:
import mlx_whisper
result = mlx_whisper.transcribe(
"audio.wav",
path_or_hf_repo="doggy8088/whisper-large-v2-cantonese-mlx",
language="zh",
)
print(result["text"])
weights.safetensors。mlx-whisper 中建議使用 --language zh(或省略讓模型自動偵測),不要指定 yue。