Automatic Speech Recognition models

Propicto
non-profit
AI & ML interests
Automatic Speech Recognition, Text Simplification, Automatic Pictogram Generation
Recent Activity
View all activity
Organization Card
PROPICTO (PRojection du langage Oral vers des unités PICTOgraphiques) is an ANR/SNF 4-year project whose main goal is to develop a line of research on Alternative and Augmented Communication (AAC) by focusing on the automatic transcription of French speech into pictogram sequences for the sake of enhancing communication access.
Collections
2
models
12

Propicto/wsd-camembert-base-semcor-wngt-fr
Text2Text Generation
•
Updated

Propicto/asr-wav2vec2-orfeo-fr
Automatic Speech Recognition
•
Updated
•
13

Propicto/asr-wav2vec2-commonvoice-15-fr
Automatic Speech Recognition
•
Updated
•
30

Propicto/t2p-nllb-200-distilled-600M-orfeo
Translation
•
Updated
•
122

Propicto/t2p-t5-large-orfeo
Translation
•
Updated
•
131

Propicto/t2p-nllb-200-distilled-600M-commonvoice
Translation
•
Updated
•
117

Propicto/t2p-nllb-200-distilled-600M-all
Translation
•
Updated
•
158

Propicto/t2p-t5-large-commonvoice
Translation
•
Updated
•
122

Propicto/t2p-nmt-orfeo
Translation
•
Updated

Propicto/t2p-nmt-commonvoice
Translation
•
Updated
datasets
None public yet