Hotpepper User Guide

Last updated 09/14/2021

Welcome to the guide for Hotpepper, an end-to-end data labeling tool set provided by Deepgram as part of its on-premise implementations. Hotpepper allows you to build custom training datasets quickly and cost effectively while maintaining data quality. It works with audio formats that can be played by the HTML5 audio player element (mp3, wav, or ogg) and improves labeling efficiency by allowing users to submit audio to a hosted or on-premise Deepgram Speech Engine for automatic speech recognition (ASR) and transcription.

With Hotpepper, you can:

  • Ensure labels are accurate and consistent
  • Monitor labeling progress by user
  • Package labeled datasets so they can be used to build models

System Requirements

As part of your on-premise implementation, your Deepgram account executive will work with you to help design your hardware layout and system configuration, which will include recommendations for Hotpepper.

To see recommended configurations, visit Configure Hardware & Software.

Guide Content

The following sections are available to help you navigate Hotpepper: