Progress in Multilingual Speech Recognition for Low Resource Languages Kurmanji Kurdish, Cree and Inuktut

no code implementations • LREC 2022 • Vishwa Gupta, Gilles Boulianne

The TDNN-F acoustic models give significantly lower WER for Kurmanji Kurdish and Cree, while BLSTM acoustic models give significantly lower WER for Inuktut.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +1

The Indigenous Languages Technology project at NRC Canada: An empowerment-oriented approach to developing language software

no code implementations • COLING 2020 • Roland Kuhn, Fineen Davis, Alain D{\'e}silets, Eric Joanis, Anna Kazantseva, Rebecca Knowles, Patrick Littell, Delaney Lothian, Aidan Pine, Caroline Running Wolf, Eddie Santos, Darlene Stewart, Gilles Boulianne, Vishwa Gupta, Brian Maracle Owennat{\'e}kha, Akwirat{\'e}kha{'} Martin, Christopher Cox, Marie-Odile Junker, Olivia Sammons, Delasie Torkornoo, Nathan Thanyeht{\'e}nhas Brinklow, Sara Child, Beno{\^\i}t Farley, David Huggins-Daines, Daisy Rosenblum, Heather Souter

This paper surveys the first, three-year phase of a project at the National Research Council of Canada that is developing software to assist Indigenous communities in Canada in preserving their languages and extending their use.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +3

Automatic Transcription Challenges for Inuktitut, a Low-Resource Polysynthetic Language

no code implementations • LREC 2020 • Vishwa Gupta, Gilles Boulianne

We introduce the first attempt at automatic speech recognition (ASR) in Inuktitut, as a representative for polysynthetic, low-resource languages, like many of the 900 Indigenous languages spoken in the Americas.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +1

Speech Transcription Challenges for Resource Constrained Indigenous Language Cree

no code implementations • LREC 2020 • Vishwa Gupta, Gilles Boulianne

With these systems and varying amounts of transcribed and text data, we show that pre-training on other languages is important for speaker-independent recognition, and even small amounts of additional text-only documents are useful.

Language Modelling speech-recognition +1