WhisperX Local Installation Kit

Description

This project enables the local installation and use of WhisperX, an advanced audio transcription system based on OpenAI's Whisper but optimized for running on local hardware with or without a GPU. This project is made possible thanks to Whisperx and Faster Whisper. This document provides a general overview of the installation and links to the website where the complete installation and usage procedure for this project can be found.

---

Requirements

• Miniconda
• CUDA (only for NVIDIA GPU users)

---

File Description

• environment-cuda.yml: Configuration file for automatic installation on systems with NVIDIA GPU.
• environment-cpu.yml: Configuration file for automatic installation on systems without NVIDIA GPU.
• app.py: Script to run the WhisperX user interface on Gradio.
• setup_environment_cuda.bat: Batch script for easy installation on Windows with NVIDIA GPU (requires CUDA 12.1 or higher).
• setup_environment_cpu.bat: Batch script for easy installation on Windows without NVIDIA GPU.
• run_script.bat: Batch script to run the application after installation.

---

Installation

For Windows Users

1. NVIDIA GPU Users (with CUDA 12.1 or higher):

If you have an NVIDIA GPU and CUDA 12.1 (or higher) installed, you can use the setup_environment_cuda.bat script for an easy installation:

1. Download the repository.
2. Double-click the setup_environment_cuda.bat file.
3. The script will:
   • Create a virtual environment.
   • Install all necessary dependencies, including GPU-optimized versions of torch and torchaudio.
4. Once the installation is complete, you can run the application by double-clicking the run_script.bat file.

2. Users without NVIDIA GPU:

If you do not have an NVIDIA GPU, use the setup_environment_cpu.bat script:

1. Download the repository.
2. Double-click the setup_environment_cpu.bat file.
3. The script will:
   • Create a virtual environment.
   • Install all necessary dependencies, including CPU-only versions of torch and torchaudio.
4. Once the installation is complete, you can run the application by double-clicking the run_script.bat file.

---

General Installation (All Platforms)

1. Miniconda: Miniconda Installation
2. CUDA: CUDA Installation (Only for NVIDIA GPU users)
3. GitHub Repository: Download and setup of the repository. See details on our website.

For a detailed step-by-step installation process, please visit our website: View complete installation process

---

Usage

To use WhisperX after completing the installation:

1. For Windows Users:

   • Simply double-click the run_script.bat file. This will:
     • Activate the virtual environment.
     • Run the Gradio user interface (app.py).

2. For Other Platforms:

   • Activate the corresponding Conda environment.
   • Run python app.py to start the Gradio user interface.

---

Changelog

[1.1.0] - 2024-08-06

• Modified default model selection:

  • For CUDA-enabled devices, changed default model from "Large-v2" to "Medium"
  • For CPU devices, kept default model as "Medium"

• Improved compute type selection:

  • For CPU devices, now uses "int8" compute type instead of "float32"
  • For CUDA devices, kept "float16" compute type

• Performance: These changes aim to balance performance and resource usage across different hardware configurations

---

Authors

• MISTER CONTENTS
• Ricardo Gonzalez

---

Languages

• Spanish
• Portuguese