Speech Recognition Models
Models for Welsh language and bilingual speech recognition
Automatic Speech Recognition • Updated • 13Note Verbatim transcription of Welsh language spontaneous speech
techiaith/whisper-large-v3-turbo-ft-btb-cv-cy
Automatic Speech Recognition • Updated • 112Note Faster, but slightly less correct, verbatim transcription of Welsh language spontaneous speech
techiaith/whisper-base-ft-btb-cv-cy-cpp
Automatic Speech Recognition • UpdatedNote whispercpp compatible model for verbatim transcription of Welsh language spontaneous speech on offline hardware such as laptops and mobile devices.
techiaith/whisper-large-v3-ft-cv-cy
Automatic Speech Recognition • Updated • 20Note Transcription of read or planned Welsh language speech
techiaith/whisper-base-ft-cv-cy-cpp
Automatic Speech Recognition • UpdatedNote whispercpp compatible model for transcription of read or planned Welsh language speech on offline hardware such as laptops and mobile devices.
techiaith/whisper-large-v3-ft-cv-cy-en
Automatic Speech Recognition • Updated • 338Note Transcription of read or planned speech in both Welsh and English
techiaith/whisper-base-ft-cv-cy-en-cpp
Automatic Speech Recognition • UpdatedNote whispercpp compatible model for transcription of Welsh and English read speech on offline hardware such as laptops and mobile devices.
techiaith/wav2vec2-btb-cv-ft-cv-cy
Automatic Speech Recognition • Updated • 3.08kNote Transcription of read or planned Welsh language speech using wav2vec2 acoustic based acoustic models and, optionally, KenLM based language models. Provides lowest WER scores.
techiaith/sacremoses-truecaser-cy
UpdatedNote A model that can correct the capitalization of speech recognition text output
techiaith/fullstop-welsh-punctuation-prediction
Token Classification • Updated • 13 • 1Note A model that can insert punctuation into speech recognition text output