Fine-tuned on Multilingual Pretrained Model CLSRIL-23. The original fairseq checkpoint is present here. When using this model, make sure that your speech input is sampled at 16kHz. Note: The result from this model is without a language model so you may witness a higher WER in some cases.

Downloads last month: 7

Model tree for Harveenchadha/vakyansh-wav2vec2-punjabi-pam-10

Finetunes

2 models

Space using Harveenchadha/vakyansh-wav2vec2-punjabi-pam-10 1

Paper for Harveenchadha/vakyansh-wav2vec2-punjabi-pam-10

CLSRIL-23: Cross Lingual Speech Representations for Indic Languages

Paper • 2107.07402 • Published Jul 15, 2021

Evaluation results

Test WER on Common Voice hi
self-reported

33.170