使用GPU一直卡住没有输出结果 #89
Description
遇到问题如下：加上-ng参数可以正常输出但是速度比较慢，不加上该参数程序卡住没有输出，所有输出如下：
nvidia@nvidia-desktop:~/code/SenseVoice.cpp/build$ ./bin/sense-voice-zcr-main -m ~/code/SenseVoice.cpp/model/sense-voice-small-fp16.gguf ~/code/SenseVoice.cpp/model/audio.wav
sense_voice_small_init_from_file_with_params_no_state: loading model from '/home/nvidia/code/SenseVoice.cpp/model/sense-voice-small-fp16.gguf'
sense_voice_init_with_params_no_state: use gpu = 1
sense_voice_init_with_params_no_state: flash attn = 0
sense_voice_init_with_params_no_state: gpu_device = 0
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
Device 0: Orin, compute capability 8.7, VMM: yes
sense_voice_init_with_params_no_state: devices = 2
sense_voice_init_with_params_no_state: backends = 2
sense_voice_model_load: version: 3
sense_voice_model_load: alignment: 32
sense_voice_model_load: data offset: 423616
sense_voice_model_load: loading model
sense_voice_model_load: n_vocab = 25055
sense_voice_model_load: n_encoder_hidden_state = 512
sense_voice_model_load: n_encoder_linear_units = 2048
sense_voice_model_load: n_encoder_attention_heads = 4
sense_voice_model_load: n_encoder_layers = 50
sense_voice_model_load: n_mels = 80
sense_voice_model_load: ftype = 1
sense_voice_model_load: vocab[25055] loaded
sense_voice_default_buffer_type: using device CUDA0 (Orin)