Akaash1
/

wav2vec-khmer-english-all

Automatic Speech Recognition

Generated from Trainer

Model card Files Files and versions

training

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on all of the podcast dataset (20hours) dataset. It achieves the following results on the evaluation set:

Wer: 0.5692

Downloads last month: 3

Safetensors

Model size

0.3B params

Tensor type

F32

·

Model tree for Akaash1/wav2vec-khmer-english-all

Base model

facebook/wav2vec2-xls-r-300m

Finetuned

(864)

this model