Custom ASR (Whisper)
Use the Whisper API with Voiceflow to simulate a custom speech recognition service
Overview
This project demonstrates Automatic Speech Recognition (ASR) using OpenAI's Whisper model running locally. It consists of a simple webpage that captures audio from the user's microphone, sends it to a custom endpoint, and displays the transcribed text and the time it took to render the transcription.
Instructions & Code Repository
You can find our open source code and step-by-step instructions on using our Dialog Manager API with it below.
https://github.com/voiceflow-gallagan/asr-demo-whisper/tree/f72b85d8f2d988c855685c13900a3f3d0a537c36
Updated about 2 months ago