Wbbbbb

Rank:

Average Model Cost: $0.0000

Number of Runs: 83,448

Models by this creator

wav2vec2-large-chinese-zh-cn

wav2vec2-large-chinese-zh-cn

wbbbbb

The wav2vec2-large-chinese-zh-cn model is a fine-tuned automatic speech recognition (ASR) model for Chinese. It is based on the XLSR-53 large model architecture and has been trained on multiple datasets including Common Voice 6.1, CSS10, and ST-CMDS. This model has been fine-tuned on an RTX3090 for 50 hours. It can be used directly for ASR tasks without the need for a separate language model. The model utilizes the HuggingSound library for usage. Evaluation results on the Chinese test data of Common Voice show the Word Error Rate (WER) and Character Error Rate (CER) of the model. The model can be cited using the provided citation information.

Read more

$-/run

83.4K

Huggingface

Similar creators