Overview / Usage

COVID 19 has made a lot of changes in the way we are working every day. Almost everyone is working from home and meetings all happen over platforms like MS Teams, Zoom, or Google meet. Marking the minutes of the meeting is a very important component in any meetings and everyone needs the same to recall and the clients also need it for the recording purpose. Someone manually always makes a note of the important points and at times, it is going wrong and is cumbersome too.

We are here to provide a solution to this challenge! Yes, upload the audio file! Sit back and relax. You will get the minutes of the meeting curated. All these happen in seconds and it's accurate too! The system is fully scalable.

Methodology / Approach

We have built a system that is able to accurately detect and transcript dialogues from audio using Audio Processing and AI Techniques. Apart from identifying dialogues, the system is able to associate voices with individuals. The user, provided with an android/Web interface, has to upload the required recording. The app routes the audio to the cloud where the algorithm processes it and results are obtained in real-time.

Technologies Used

Faster RCNN, Speech Segmentation, and Speech Diarization.

Repository

https://drive.google.com/drive/folders/14w8VrzyYVGhFVz5fRPSy8sDh8j0uJinr?usp=sharing

Collaborators