Script AI is an innovative tool designed to enhance public speaking and acting performances. Utilizing the power of artificial intelligence, Script AI provides real-time feedback on speech delivery by analyzing video recordings. Ideal for actors and public speakers, this tool refines your presentation skills for maximum audience impact.
- Video Recording: Capture your performance directly through the platform for analysis.
- Speech Analysis: Advanced algorithms offer feedback on speech clarity, tempo, and emotion.
- Script Processing: Input scripts via text or upload PDFs for seamless integration.
- Real-time Feedback: Gain insights on improving delivery, tone, and audience engagement.
- User-friendly Interface: Navigate through the recording and analysis process with ease.
- Frontend: Crafted with modern web technologies for a responsive user experience.
- Backend: Built on Flask, our robust backend handles complex audio and video processing tasks efficiently.
- Data Storage: Utilizes AWS S3 for secure and scalable storage of video and script data.
- Machine Learning: Employs GPT Vision for facial expression analysis and Together API for generating the feedback.
Ensure you have the following installed:
- Python 3.6 or later
- Flask
- boto3
- openai-whisper (for audio analysis)
- All dependencies in
requirements.txt
- Clone the repository: git clone
- Install dependencies: pip install -r requirements.txt
- Configure AWS and OpenAI credentials in your environment or
.envfile.
Execute the following command:
Then, visit http://localhost:5003 in your browser.
- Text: Paste script text on the homepage.
- PDF: Drag and drop or select a PDF script file.
Hit the record button post-upload and grant browser camera and microphone access.
Post-performance, Script AI processes the video and audio, providing detailed speech insights.
Here's Script AI in action:
We welcome contributions! Feel free to submit pull requests or open issues for discussion.
Script AI is released under the MIT License.