A versatile tool for multi-language audio transcription, transliteration, and word timestamp extraction. Supporting English, Telugu, Hindi, Tamil, Malayalam, and Kannada, it offers a Gradio-powered interface for easy audio analysis, script conversion, and precise word tracking. Perfect for linguists, researchers, and developers.
- Multi-Language Transcription: Supports transcription in English, Telugu, Hindi, Tamil, Malayalam, and Kannada.
- Script Transliteration: Transliterates the transcribed text into a different script for enhanced compatibility.
- Word Timestamp Extraction: Extracts timestamps for specific words within the audio file.
- User-Friendly Interface: A Gradio-based web interface for easy interaction and instant results.
- Audio Format Compatibility: Converts uploaded audio files to WAV format for seamless processing.
To get started , follow these steps:
-
Clone the Repository:
git clone https://github.com/LSShrivathsan/Transcribe-Transliterate-Timestamp-Audiofile.git cd Transcribe-Transliterate-Timestamp-Audiofile
-
Install the Required Dependencies:
pip install -r requirements.txt
-
Run the Gradio Interface:
python app.py
- Upload Audio: Use the interface to upload an audio file.
- Select Language: Choose the transcription language from the options.
- Search Word: Enter a word to extract its timestamps from the audio.
- View Results: See the transcribed text, transliterated script and word timestamps.
Here's how to use the tool:
-
Upload an Audio File:
- The app supports various audio formats. Upload your file to start processing.
-
Choose a Language:
- Select the desired language for transcription.
-
Enter a Word for Timestamp Extraction:
- Input the word you want to search within the audio. The app will provide timestamps where this word occurs.
-
View and Download Results:
- Get the transcription, transliteration, and timestamps. You can copy or download the results.
We welcome contributions! If you'd like to contribute, please fork the repository and use a feature branch. Pull requests are warmly welcome.
This project is licensed under the MIT License
For any questions or inquiries, please contact - [email protected]