Merge pull request #129 from chenqianhe/main

Add `large-v3-turbo`

.github/workflows/base.yml

@@ -12,8 +12,8 @@ jobs:
     strategy:
       matrix:
         python-version: ['3.9', '3.10']
-        #  macos did not support m1 for now
-        os: [ubuntu, windows, macos]
+        #  Wait for fix on macos-m1: https://github.com/federicocarboni/setup-ffmpeg/issues/21
+        os: [ubuntu, windows, macos-12]
     steps:
       - uses: actions/checkout@v3
       - name: Set up Python ${{ matrix.python-version }}
@@ -32,7 +32,7 @@ jobs:
           path: ${{ steps.set_variables.outputs.PIP_CACHE }}
           key: ${{ runner.os }}-pip-${{ steps.set_variables.outputs.PY }}
 
-      - name: Setup ffmpeg for differnt platforms 
+      - name: Setup ffmpeg for different platforms
         uses: FedericoCarboni/setup-ffmpeg@v3
 
       - name: Install dependencies

README.md

@@ -2,6 +2,12 @@
 
 AutoCut 对你的视频自动生成字幕。然后你选择需要保留的句子，AutoCut 将对你视频中对应的片段裁切并保存。你无需使用视频编辑软件，只需要编辑文本文件即可完成剪切。
 
+**2024.10.05更新**：支持 `large-v3-turbo` [模型](https://github.com/openai/whisper/discussions/2363)，提供更快的转录速度。
+
+```shell
+autocut -t xxx --whisper-model large-v3-turbo
+````
+
 **2024.03.10更新**：支持 pip 安装和提供 import 转录相关的功能
 
 ```shell

autocut/__init__.py

@@ -1,4 +1,4 @@
-__version__ = "1.0.2"
+__version__ = "1.1.0"
 
 from .type import LANG, WhisperModel, WhisperMode
 from .utils import load_audio

autocut/type.py

@@ -72,6 +72,7 @@ class WhisperModel(Enum):
     LARGE = "large"
     LARGE_V2 = "large-v2"
     LARGE_V3 = "large-v3"
+    LARGE_V3_TURBO = "large-v3-turbo"
 
     @staticmethod
     def get_values():