Your Ultimate Developer AI for transcribing
Tired of manually transcribing audio and video files? This project brings you a streamlined, powerful, and visually appealing solution using the cutting-edge AssemblyAI API. Transform your audio and video content into easily searchable text with advanced analysis features, all within a user-friendly Streamlit interface.
- Effortless File Uploads: Drag and drop your audio (mp3, wav, ogg, m4a) and video (mp4) files for instant transcription.
- Social Media Integration: Seamlessly transcribe YouTube videos and Instagram Reels directly from their URLs.
- Intelligent Analysis: Unlock deeper insights with speaker diarization, entity detection, and key phrase extraction. Visualize your data with interactive charts!
- Instant Downloads: Download your perfectly formatted transcripts as text files for easy access and sharing.
- Intuitive Interface: Enjoy a smooth and efficient user experience thanks to the elegant Streamlit framework.
This project leverages the power of:
- Streamlit: The go-to framework for building interactive web apps in Python.
- AssemblyAI: The leading speech-to-text API, providing accurate and fast transcriptions.
- MoviePy: Handles video processing, ensuring smooth audio extraction.
- yt-dlp: Efficiently downloads videos from YouTube and Instagram.
- Plotly: Creates stunning interactive charts for data visualization.
- Pandas: Simplifies data manipulation for a seamless user experience.
- Clone the Repository:
git clone https://github.com/saoud30/Audio-AI.git
- Install Dependencies:
pip install -r requirements.txt
- Configure API Key:
- Create a
.env
file in the project root. - Add your AssemblyAI API key:
ASSEMBLY_AI_API_KEY=<your_api_key>
- Create a
- Run the App:
streamlit run app.py
- Ensure you have
streamlit
and all dependencies installed. - Obtain your AssemblyAI API key from their website and replace
<your_api_key>
. - Processing time may vary depending on the length of your audio/video files.
- For Instagram Reels, ensure your browser's cookies are properly configured (Chrome cookies are recommended).
We welcome contributions! Please feel free to open issues or submit pull requests.
This project is licensed under the MIT License - see the LICENSE file for details.
- GitHub: @Saoud30
- Email: mohdhashimansari30@gmail.com
- Twitter: @Shazyansar
Made with ❤️ by Shazy