transcribe model

/overview

an ai-powered application that performs speech transcription and speaker segmentation from uploaded audio or video files. the system extracts audio from media, segments the speech by speaker, and generates transcriptions through a trained model.

/implementation

this project utilizes a fastapi backend for processing and inference, connected to a react frontend for user interaction. the ai speech model handles transcription, while automatic video-to-audio conversion streamlines the pipeline.

/the tech stack