Omnilingual ASR: Advancing automatic speech recognition for 1600 languages
1–10 of 48 posts
Re: Omnilingual ASR: Advancing automatic speech recognition for 1600 languages
#2Re: Omnilingual ASR: Advancing automatic speech recognition for 1600 languages
#3any insights on latency?
Re: Omnilingual ASR: Advancing automatic speech recognition for 1600 languages
#4HF Demo: https://huggingface.co/spaces/facebook/omniasr-transcription... GitHub: https://github.com/facebookresearch/omnilingual-asr
Thanks! I've added those links to the toptext as well.
Re: Omnilingual ASR: Advancing automatic speech recognition for 1600 languages
#5How hard is it to make TTS out of this? A few independent journalists from Belarus asked for TTS in their language, but I am no expert, was thinking about re-using Mozilla's work. What's the easiest way to get working TTS for a language?
Re: Omnilingual ASR: Advancing automatic speech recognition for 1600 languages
#6This seems like a massive improvement for openly available local ASR. Even the 300M model outperforms whisper-large-v3 according to the paper's benchmarks.
Re: Omnilingual ASR: Advancing automatic speech recognition for 1600 languages
#7How hard is it to make TTS out of this? A few independent journalists from Belarus asked for TTS in their language, but I am no expert, was thinking about re-using Mozilla's work. What's the easiest way to get working TTS for a language?
From TFA, it says that it’s extremely easy to add new languages with just a few examples. I didn’t see specifics on how “few” it really is, though.
Re: Omnilingual ASR: Advancing automatic speech recognition for 1600 languages
#8the global language explorer is fascinating -great work guys
https://aidemos.atmeta.com/omnilingualasr/language-globe
- we are getting closer to BabelFish.. at least for the Earth!
Re: Omnilingual ASR: Advancing automatic speech recognition for 1600 languages
#9Only a few gb of weights will recognize speech in 1600+ languages.
Freely downloadable and usable by anyone for almost anything.
We truly live in the future.
Re: Omnilingual ASR: Advancing automatic speech recognition for 1600 languages
#10How hard is it to make TTS out of this? A few independent journalists from Belarus asked for TTS in their language, but I am no expert, was thinking about re-using Mozilla's work. What's the easiest way to get working TTS for a language?
From TFA, it says that it’s extremely easy to add new languages with just a few examples. I didn’t see specifics on how “few” it really is, though.
This is ASR not TTS though.