 |
Conference papers
Plugging a neural phoneme recognizer into a simple language model: a workflow for low-resource settings
Plugging a neural phoneme recognizer into a simple language model: a workflow for low-resource settings
Abstract : Recently, several works have shown that fine-tuning a multilingual model of speech representation (typically XLS-R) with very small amounts of annotated data allows for the development of phonemic transcription systems of sufficient quality to help field linguists in their efforts to document the languages of the world. In this work, we explain how the quality of these systems can be improved by a very simple method, namely integrating them with a language model. Our experiments on an endangered language, Japhug (Trans-Himalayan/Tibeto-Burman), show that this approach can significantly reduce the WER, reaching the stage of automatic recognition of entire words.
https://halshs.archives-ouvertes.fr/halshs-03625581
Contributor : Alexis Michaud Connect in order to contact the contributor
Submitted on : Thursday, March 31, 2022 - 5:51:19 AM
Last modification on : Friday, August 5, 2022 - 11:58:04 AM
Long-term archiving on: : Friday, July 1, 2022 - 7:08:31 PM
Contributor : Alexis Michaud Connect in order to contact the contributor
Submitted on : Thursday, March 31, 2022 - 5:51:19 AM
Last modification on : Friday, August 5, 2022 - 11:58:04 AM
Long-term archiving on: : Friday, July 1, 2022 - 7:08:31 PM
File
Interspeech2022_ASR_Endangered...
Files produced by the author(s)
Licence
Distributed under a Creative Commons Attribution - NonCommercial - ShareAlike 4.0 International License
