TingChen-ppmc
/

whisper-small-Changsha

Automatic Speech Recognition

Model card Files Files and versions

This is a finetuned version of Whisper-small on Changsha_Dialect_Conversational_Speech_Corpus

Results

CER: 37.77

WER: 66.84

Chinese Word Error Rate (WER) is based on Jieba Word segmentation

Downloads last month: 21

Safetensors

Model size

0.2B params

Tensor type

F32

·

Model tree for TingChen-ppmc/whisper-small-Changsha

Base model

openai/whisper-small

Finetuned

(3817)

this model

Dataset used to train TingChen-ppmc/whisper-small-Changsha