Speechmatics Adds French Canadian and Brazilian Portuguese
Speechmatics, a speech recognition technology provider, has added support for French Canadian and Brazilian Portuguese in its global language packs.
For the Canadian French language pack, Speechmatics trained its engine on 1,100 hours of Canadian French voice data, drawing from publicly available sources, including Common Voice and a Canadian French production house. This training has led to an average relative reduction of 18 percent in word error rate (WER) for its standard model and 13 percent for its advanced model.
Meanwhile, Brazilian Portuguese was trained using Common Voice, as well as multilingual Librispeech. This additional training has led to overall improvements in the general Portuguese language pack and a reduction in WER of 5 percent for standard models and 3 percent for advanced models respectively.
"At Speechmatics, we are striving to understand every voice, regardless of race, gender, or accent. We have been working tirelessly to improve and simplify our machine learning pipelines to ingest and process new data rapidly. Now we have the flexibility to adapt to events as they occur," said John Hughes, accuracy team lead at Speechmatics, in a statement.
The conmpany chose French Canadian and Brazilian Portuguese "due to the vast number of speakers worldwide and the lack of representative training data," Hughes said. "Now we have a clear advantage in accuracy over all our competitors: Our models, powered by self-supervised learning, continue to move us closer to making speech recognition valuable to everyone, everywhere, anytime."