Hubro Documentation Help

Audio transcription

Voice recognition light

Audio transcription is a component that can significantly facilitate processing of data collected in focus group meetings or various in-person interviews.

Hubro user can use audio transcription in two different ways

  1. The user has just finished audio recording using Hubro, and would like to use audio transcription to transcribe the recording into text

  2. The user owns an .mp3 file that can be uploaded to the system in order to initiate the scheduling of audio transcription task

The voice recognition engine being used is Whisper. Whisper is seamlessly integrated into the architecture orchestrated by Kubernetes. In practice, this approach ensures horizontal scalability of the Whisper service and enables parallel processing of requests.

Last modified: 23 September 2024