Skip to content

Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.

update offline-stream

雾聪c76c6da543fe
diff --git a/runtime/onnxruntime/src/offline-stream.cpp b/runtime/onnxruntime/src/offline-stream.cpp
index abe13aee5..9a3b07ae0 100644
--- a/runtime/onnxruntime/src/offline-stream.cpp
+++ b/runtime/onnxruntime/src/offline-stream.cpp
@@ -38,6 +38,10 @@ OfflineStream::OfflineStream(std::map<std::string, std::string>& model_path, int
if(use_gpu){
#ifdef USE_GPU
asr_handle = make_unique<ParaformerTorch>();
+ #else
+ LOG(ERROR) <<"GPU is not supported! CPU will be used! If you want to use GPU, please add -DGPU=ON when cmake";
+ asr_handle = make_unique<Paraformer>();
+ use_gpu = false;
#endif
}else{
asr_handle = make_unique<Paraformer>();