Speech to text

This plugin provides tools to perform speech to text (STT) using Deep Learning.
Speech to text consists in automatic transcription of an audio speech.
This plugin comes with four recipes and a macro.

Plugin information

Version0.1.0
AuthorDataiku Labs (Arthur Douillard)
Released2018-08-01
Last updated2018-08-01
LicenseMPL-2.0

How To Use

The recipe takes in input DeepSpeech weights and a folder where audio files of the format .wav are stored. The recipe's result is a dataset with the audio filename and the associated transcription.