User-friendly automatic transcription of low-resource languages: Plugging ESPnet into Elpis

TL;DR

This paper details the integration of ESPnet into Elpis to facilitate user-friendly, end-to-end speech recognition for low-resource languages through a graphical interface, including new recipes and UI improvements.

Contribution

It introduces a new integration of ESPnet into Elpis, enabling easier access to speech recognition models for language workers, with UI enhancements and support for new datasets.

Findings

Successful development of an ESPnet recipe in Elpis

Preliminary results on multiple datasets, including new ones

Enhanced user interface and Docker support for easier deployment

Abstract

This paper reports on progress integrating the speech recognition toolkit ESPnet into Elpis, a web front-end originally designed to provide access to the Kaldi automatic speech recognition toolkit. The goal of this work is to make end-to-end speech recognition models available to language workers via a user-friendly graphical interface. Encouraging results are reported on (i) development of an ESPnet recipe for use in Elpis, with preliminary results on data sets previously used for training acoustic models with the Persephone toolkit along with a new data set that had not previously been used in speech recognition, and (ii) incorporating ESPnet into Elpis along with UI enhancements and a CUDA-supported Dockerfile.