Update README.md

README.md

@@ -1,38 +1,30 @@
----
-license: cc-by-nc-sa-4.0
-language:
-- en
-pretty_name: r
----
-# Speaker Diarization and Identification Toolkit 
+# Speaker Diarization and Identification Toolkit
 
-Welcome to the **Speaker Diarization and Identification Toolkit**, a suite of tools for audio processing and speaker analysis. This repository offers user-friendly Python scripts for extracting audio, performing speaker diarization, and interactively identifying specific speakers from multi-speaker recordings. 
+Welcome to the **Speaker Diarization and Identification Toolkit**, a suite of tools for processing audio and identifying speakers. This repository offers Python scripts for extracting audio, performing speaker diarization, identifying speakers, and trimming segments of interest.
 
 ---
 
-## 🌟 Features
+## Features
 
-- **Speaker Diarization**: Automatically segment and label audio by speaker using the Hugging Face `pyannote.audio` pipeline.
-- **Audio Extraction**: Seamlessly extract audio tracks from video files and convert them to WAV format.
-- **Speaker Identification**: Interactively map speakers in multi-speaker environments using diarization data and audio playback.
-- **Rich User Interface**: Leveraging the `rich` library for visually appealing and intuitive prompts, error messages, and feedback.
+- **Speaker Diarization**: Automatically segment and label audio by speaker.
+- **Audio Extraction**: Extract audio tracks from video files.
+- **Speaker Identification**: Interactively map speakers using diarization data.
+- **Segment Trimming**: Extract and trim speaker-specific audio segments.
 
 ---
 
-## 🔧 Requirements
-
-To get started, ensure you have the following:
+## Requirements
 
-- **Python**: Version 3.8 or later.
-- **Hugging Face Token**: A personal access token to use the `pyannote.audio` pipeline.
-- **System Tools**: `ffmpeg` installed on your system.
-- **Python Libraries**:
+- Python 3.8+
+- Hugging Face token for `pyannote.audio`
+- System dependency: `ffmpeg`
+- Python libraries:
   - `rich`
   - `pandas`
   - `pyannote.audio`
   - `playsound`
 
-Install all dependencies with:
+Install dependencies with:
 
 ```bash
 pip install -r requirements.txt
@@ -40,57 +32,36 @@ pip install -r requirements.txt
 
 ---
 
-## 🚀 Setup
-
-### Step 1: Clone the Repository
-
-Clone this repository to your local machine:
-
-```bash
-git clone https://huggingface.co/datasets/ScottishHaze/diarization-project
-cd diarization-project
-```
-
-### Step 2: Configure the Environment
-
-1. **Install `ffmpeg`**:
-   - **Linux**: `sudo apt install ffmpeg`
-   - **MacOS**: `brew install ffmpeg`
-   - **Windows**: Download from [ffmpeg.org](https://ffmpeg.org/).
-
-2. **Set Up a Hugging Face Token**:
-   - Create an account at [huggingface.co](https://huggingface.co/).
-   - Generate a token from your account settings.
+## Setup
 
-3. **Environment Configuration**:
-   - Create an `.env` file in the project root and add:
-
-     ```env
-     HF_TOKEN=your_hugging_face_token
-     ```
-
-### Step 3: Organize Files
-
-Use the following directory structure or customize paths interactively when running the scripts:
+### Directory Structure
 
 ```plaintext
 project/
-├── diarization # JSON files will go here
-├── extracted-audio # WAV files will go here
-├── scripts
+├── input/
+│   ├── video/       # Video files
+│   ├── audio/       # WAV files from videos
+├── output/
+│   ├── json/        # Diarization results
+│   ├── audio/       # Cropped and trimmed speaker segments
+├── scripts/
 │   ├── ffmpeg-extract-audio-from-video.py
 │   ├── diarization-extract-speaker-info-from-wav.py
-│   └── identify-single-speaker.py
-├── miniconda # working program/dir, !!! may not appear on your system, and that's fine !!!
+│   ├── identify-single-speaker.py
+│   ├── speaker-crop-and-trim.py
+├── data/
+│   └── mappings.csv # Speaker mappings
 ```
 
+Ensure the directory structure matches the expected inputs and outputs for each script.
+
 ---
 
-## 🎬 Usage
+## Usage
 
 ### 1. Extract Audio from Video
 
-Use `ffmpeg-extract-audio-from-video.py` to extract English audio tracks from video files:
+Run the `ffmpeg-extract-audio-from-video.py` script to extract audio tracks from video files:
 
 ```bash
 python ffmpeg-extract-audio-from-video.py
@@ -98,7 +69,7 @@ python ffmpeg-extract-audio-from-video.py
 
 ### 2. Perform Speaker Diarization
 
-Run `diarization-extract-speaker-info-from-wav.py` to analyze WAV files and generate speaker segmentation JSON files:
+Run the `diarization-extract-speaker-info-from-wav.py` script to analyze WAV files and generate speaker segmentation JSON files:
 
 ```bash
 python diarization-extract-speaker-info-from-wav.py
@@ -106,38 +77,32 @@ python diarization-extract-speaker-info-from-wav.py
 
 ### 3. Identify Target Speaker
 
-Use `identify-single-speaker.py` to interactively identify target speakers:
+Use the `identify-single-speaker.py` script to interactively identify and map target speakers. For multiple parts / TV series
+it is recommended you listen to multiple segments of each speaker before identifying "Y". Sometimes after diarization, the targeted
+speaker can be blended in with other speakers.
 
 ```bash
 python identify-single-speaker.py
 ```
-  When doing series/multipart analysis, it is recommended you listen to a few sections before saying YES to a speaker.
-
----
 
-## 🤝 Contributing
+### 4. Crop and Trim Speaker Segments
 
-Contributions are highly appreciated! To contribute:
+Run the `speaker-crop-and-trim.py` script to extract, crop, and trim audio segments for identified speakers:
 
-1. Fork this repository.
-2. Create a branch for your feature or bug fix.
-3. Submit a pull request with a detailed description of changes. 
+```bash
+python speaker-crop-and-trim.py
+```
 
 ---
 
-## 📜 License
+## License
 
-This project is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) License. Refer to the [LICENSE](LICENSE) file for full details of the terms. 
+This project is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0). Refer to the [LICENSE](LICENSE) file for details.
 
 ---
 
-## 🌍 Acknowledgments
-
-Gratitude goes to:
+## Acknowledgments
 
-- [Hugging Face](https://huggingface.co/) for providing robust pre-trained models.
-- The creators of `rich`, `pyannote.audio`, and `ffmpeg` for their indispensable tools. 
-
----
+Gratitude to [Hugging Face](https://huggingface.co/) and the developers of `rich`, `pyannote.audio`, and `ffmpeg` for their invaluable tools.
 
-Thank you for exploring this toolkit. If you encounter any issues or have suggestions, feel free to open an issue or contribute!
+Thank you for exploring this toolkit!