Dolphin ASR

A multilingual, multitask ASR model supporting 40 Eastern languages and 22 Chinese dialects.

This model is from DataoceanAI/Dolphin, for speech recognition in Eastern languages including Chinese, Japanese, Korean, and many more.

Usage Notes

• The model supports 40 Eastern languages and 22 Chinese dialects
• You can let the model auto-detect language or specify language and region
• Timestamps can be included in the output
• Speech can be padded to 30 seconds for better processing

Credits

• Model: DataoceanAI/Dolphin
• Paper: Dolphin: A Multilingual Model for Eastern Languages