fix(paraformer): keep one keyed result per utterance when no tokens are predicted - #3797

fix(paraformer): keep one keyed result per utterance when no tokens are predicted#3797

LauraGPT merged 1 commit into

modelscope:mainmodelscope/FunASR:mainfrom

MohammadHijjawi97:fix/paraformer-empty-token-batchMohammadHijjawi97/FunASR:fix/paraformer-empty-token-batchCopy head branch name to clipboard