Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
Python asraudiochineseemotion-recognitionfunasrmcp-servermultilingual-asropenai-compatible-apiparaformerpunctuationpytorchreal-time-asrspeaker-diarizationspeech-recognitionspeech-to-textstreaming-asrtranscriptionvllmvoice-activity-detectionwhisper-alternative
fix tpass_stream
雾聪6bbbf14c0080
diff --git a/runtime/onnxruntime/src/funasrruntime.cpp b/runtime/onnxruntime/src/funasrruntime.cppindex 5e2aa935e..2846930e7 100644--- a/runtime/onnxruntime/src/funasrruntime.cpp+++ b/runtime/onnxruntime/src/funasrruntime.cpp@@ -797,7 +797,7 @@return mm;}#ifdef USE_GPU- auto paraformer_torch = dynamic_cast<funasr::ParaformerTorch*>(offline_stream->asr_handle.get());+ auto paraformer_torch = dynamic_cast<funasr::ParaformerTorch*>(tpass_stream->asr_handle.get());if(paraformer_torch !=nullptr){if (paraformer_torch->lm_){mm = new funasr::WfstDecoder(paraformer_torch->lm_.get(),