Faster Whisper GUI

Faster Whisper GUI is a Python-based graphical user interface for the faster-whisper library, providing an easy way to generate transcripts from audio files.

Enhancements

This fork includes several enhancements over the original version:

1. Model Download Mirror Support

• Added support for using https://hf-mirror.com as a model download mirror
• Toggle option in the UI to enable/disable mirror usage
• Models are downloaded to a local models/ directory for better management

2. Improved File Encoding Handling

• Fixed issues with non-ASCII characters in file paths (e.g. Chinese characters) in linux
• Implemented temporary file copying mechanism for problematic file paths
• Better error handling and reporting for encoding issues

3. Test

• Only tested on Debian 13 with cpu infenrence，install requirements packages via pip w mirror (pixi is unavailable because due to the mirror's slow response time and lack of conda-forge packages)

4. File tree

.
├── faster_whisper_gui.py
├── FastWhisperGui.ui
├── icon.ico
├── icon.png
├── LICENSE
├── make_app.cmd
├── make_app _cpu.cmd
├── models
│   └── models--Systran--faster-whisper-large-v2
│       ├── blobs
│       │   ├── 346bb8684801a61c15ce82bcb80c29a63c2d1f94
│       │   ├── 7818adb6de9fa3064d3ff81226fdd675be1f6344
│       │   ├── bf2a9746382e1aa7ffff6b3a0d137ed9edbd9670c3b87e5d35f5e85e70d0333a
│       │   └── c9074644d9d1205686f16d411564729461324b75
│       ├── refs
│       │   └── main
│       └── snapshots
│           └── f0fe81560cb8b68660e564f55dd99207059c092e
│               ├── config.json -> ../../blobs/346bb8684801a61c15ce82bcb80c29a63c2d1f94
│               ├── model.bin -> ../../blobs/bf2a9746382e1aa7ffff6b3a0d137ed9edbd9670c3b87e5d35f5e85e70d0333a
│               ├── tokenizer.json -> ../../blobs/7818adb6de9fa3064d3ff81226fdd675be1f6344
│               └── vocabulary.txt -> ../../blobs/c9074644d9d1205686f16d411564729461324b75
├── pixi.lock
├── pixi.toml
├── README.md
└── requirements.txt

Installation

1. Clone the repository:

   git clone https://github.com/Velkzilla/faster-whisper-gui.git
   cd faster-whisper-gui

2. Install the required dependencies:

   pip install -r requirement.txt  -i https://mirror.sjtu.edu.cn/pypi/web/simple

3. Run the application:

   python faster-whisper-gui.py

Building

Added linux make script

pip install pyinstaller -i https://mirror.sjtu.edu.cn/pypi/web/simple
./make_app_cpu.sh 

Features

• Simple and intuitive interface.
• Choose your input file and output folder for transcripts.
• Select model size for transcription accuracy.
• GPU acceleration support for faster processing.

Download

Download the latest version of Faster Whisper GUI from the Releases section.

How to Use

Step 1: Choose Your File

Select the audio file you want to transcribe using the file picker.

Step 2: Choose the Transcript Folder

Pick the folder where you want the generated transcript to be saved.

Step 3: Choose the Model Size

Select the model size for transcription. Larger models are more accurate but may require more processing power.

Step 4: Select GPU or CPU Mode

• GPU Mode: Choose this option if you have a GPU and want faster transcription.
• CPU Mode: Use this if you do not have a GPU or prefer CPU processing.

Step 5: Start or Stop Transcription

Click the "Transcript" button to start the transcription process. You can stop it at any time by clicking the "Stop" button.

Requirements

• Python 3.9+
• Dependencies listed in requirements.txt
• Optional: GPU with CUDA support for GPU acceleration

License

This project is licensed under the MIT License. See the LICENSE file for details.

Acknowledgments

This application is built on top of the faster-whisper library.

---

Feel free to contribute to the project or report any issues you encounter!