sherpa-onnx
sherpa-onnx copied to clipboard
Published
20 hours ago •
k2-fsa
k2-fsa
go-api-examples/vad-asr-paraformer bugs
Example: sherpa-onnx/go-api-examples/vad-asr-paraformer
go version go1.22.0 windows/amd64
bug info:
2024/04/02 14:06:59.904411 Selected default input device: 麦克风阵列 (适用于数字麦克风的英特尔® 智音技术)
2024/04/02 14:07:00.273700 Started! Please speak
2024/04/02 14:07:18.407361 Detected speech
2024/04/02 14:07:19.597375 Duration: 1.03 seconds
Exception 0xc0000005 0x0 0x1fd95e5c000 0x7ff91c7a3520
PC=0x7ff91c7a3520
signal arrived during external code execution
runtime.cgocall(0xca8090, 0xc00006be10)
D:/opt/Go/src/runtime/cgocall.go:157 +0x3e fp=0xc00006bde8 sp=0xc00006bdb0 pc=0xc0585e
github.com/k2-fsa/sherpa-onnx-go-windows._Cfunc_DecodeOfflineStream(0x1fd66019060, 0x1fd76a2e840)
_cgo_gotypes.go:505 +0x4d fp=0xc00006be10 sp=0xc00006bde8 pc=0xc9eb2d
github.com/k2-fsa/sherpa-onnx-go-windows.(*OfflineRecognizer).Decode.func1(0x4060?, 0xc00000a610?)
C:/Users/Administrator/go/pkg/mod/github.com/k2-fsa/[email protected]/sherpa_onnx.go:493 +0x73 fp=0xc00006be60 sp=0xc00006be10 pc=0xca0cd3
github.com/k2-fsa/sherpa-onnx-go-windows.(*OfflineRecognizer).Decode(0xc00000a0c0?, 0x0?)
C:/Users/Administrator/go/pkg/mod/github.com/k2-fsa/[email protected]/sherpa_onnx.go:493 +0x13 fp=0xc00006be80 sp=0xc00006be60 pc=0xca0c33
main.decode(0xc00000a0c0, 0xc000180020, 0x0)
D:/Desktop/sherpa-onnx-master/go-api-examples/vad-asr-paraformer/main.go:127 +0xd9 fp=0xc00006bfb8 sp=0xc00006be80 pc=0xca6e99
main.main.gowrap5()
D:/Desktop/sherpa-onnx-master/go-api-examples/vad-asr-paraformer/main.go:112 +0x28 fp=0xc00006bfe0 sp=0xc00006bfb8 pc=0xca6b48
runtime.goexit({})
D:/opt/Go/src/runtime/asm_amd64.s:1695 +0x1 fp=0xc00006bfe8 sp=0xc00006bfe0 pc=0xc65781
created by main.main in goroutine 1
D:/Desktop/sherpa-onnx-master/go-api-examples/vad-asr-paraformer/main.go:112 +0x730
goroutine 1 gp=0xc000058000 m=0 mp=0xd8de00 [syscall]:
runtime.cgocall(0xca7930, 0xc000117cb0)
D:/opt/Go/src/runtime/cgocall.go:157 +0x3e fp=0xc000117c88 sp=0xc000117c50 pc=0xc0585e
github.com/gordonklaus/portaudio._Cfunc_Pa_ReadStream(0x1fd65702de0, 0xc00012a000, 0x640)
_cgo_gotypes.go:521 +0x55 fp=0xc000117cb0 sp=0xc000117c88 pc=0xc9b135
github.com/gordonklaus/portaudio.(*Stream).Read.func1(0xc000117d00?, 0xc00012a000, 0x640)
C:/Users/Administrator/go/pkg/mod/github.com/gordonklaus/[email protected]/portaudio.go:1004 +0x6e fp=0xc000117ce8 sp=0xc000117cb0 pc=0xc9e14e
github.com/gordonklaus/portaudio.(*Stream).Read(0xc000132000)
C:/Users/Administrator/go/pkg/mod/github.com/gordonklaus/[email protected]/portaudio.go:1004 +0x51 fp=0xc000117d10 sp=0xc000117ce8 pc=0xc9e091
main.main()
D:/Desktop/sherpa-onnx-master/go-api-examples/vad-asr-paraformer/main.go:89 +0x610 fp=0xc000117f50 sp=0xc000117d10 pc=0xca6750
runtime.main()
D:/opt/Go/src/runtime/proc.go:271 +0x28b fp=0xc000117fe0 sp=0xc000117f50 pc=0xc39bab
runtime.goexit({})
D:/opt/Go/src/runtime/asm_amd64.s:1695 +0x1 fp=0xc000117fe8 sp=0xc000117fe0 pc=0xc65781
goroutine 2 gp=0xc000058700 m=nil [force gc (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
D:/opt/Go/src/runtime/proc.go:402 +0xce fp=0xc00005bfa8 sp=0xc00005bf88 pc=0xc39fae
runtime.goparkunlock(...)
D:/opt/Go/src/runtime/proc.go:408
runtime.forcegchelper()
D:/opt/Go/src/runtime/proc.go:326 +0xb8 fp=0xc00005bfe0 sp=0xc00005bfa8 pc=0xc39e38
runtime.goexit({})
D:/opt/Go/src/runtime/asm_amd64.s:1695 +0x1 fp=0xc00005bfe8 sp=0xc00005bfe0 pc=0xc65781
created by runtime.init.6 in goroutine 1
D:/opt/Go/src/runtime/proc.go:314 +0x1a
goroutine 3 gp=0xc000058a80 m=nil [GC sweep wait]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
D:/opt/Go/src/runtime/proc.go:402 +0xce fp=0xc00005df80 sp=0xc00005df60 pc=0xc39fae
runtime.goparkunlock(...)
D:/opt/Go/src/runtime/proc.go:408
runtime.bgsweep(0xc000068000)
D:/opt/Go/src/runtime/mgcsweep.go:278 +0x94 fp=0xc00005dfc8 sp=0xc00005df80 pc=0xc249b4
runtime.gcenable.gowrap1()
D:/opt/Go/src/runtime/mgc.go:203 +0x25 fp=0xc00005dfe0 sp=0xc00005dfc8 pc=0xc192c5
runtime.goexit({})
D:/opt/Go/src/runtime/asm_amd64.s:1695 +0x1 fp=0xc00005dfe8 sp=0xc00005dfe0 pc=0xc65781
created by runtime.gcenable in goroutine 1
D:/opt/Go/src/runtime/mgc.go:203 +0x66
goroutine 4 gp=0xc000058c40 m=nil [GC scavenge wait]:
runtime.gopark(0xc000068000?, 0xd02970?, 0x1?, 0x0?, 0xc000058c40?)
D:/opt/Go/src/runtime/proc.go:402 +0xce fp=0xc00006ff78 sp=0xc00006ff58 pc=0xc39fae
runtime.goparkunlock(...)
D:/opt/Go/src/runtime/proc.go:408
runtime.(*scavengerState).park(0xd8d7a0)
D:/opt/Go/src/runtime/mgcscavenge.go:425 +0x49 fp=0xc00006ffa8 sp=0xc00006ff78 pc=0xc223a9
runtime.bgscavenge(0xc000068000)
D:/opt/Go/src/runtime/mgcscavenge.go:653 +0x3c fp=0xc00006ffc8 sp=0xc00006ffa8 pc=0xc2293c
runtime.gcenable.gowrap2()
D:/opt/Go/src/runtime/mgc.go:204 +0x25 fp=0xc00006ffe0 sp=0xc00006ffc8 pc=0xc19265
runtime.goexit({})
D:/opt/Go/src/runtime/asm_amd64.s:1695 +0x1 fp=0xc00006ffe8 sp=0xc00006ffe0 pc=0xc65781
created by runtime.gcenable in goroutine 1
D:/opt/Go/src/runtime/mgc.go:204 +0xa5
goroutine 5 gp=0xc000059180 m=nil [finalizer wait]:
runtime.gopark(0xc00005fe48?, 0xc0f0c5?, 0xa8?, 0x1?, 0xc000058000?)
D:/opt/Go/src/runtime/proc.go:402 +0xce fp=0xc00005fe20 sp=0xc00005fe00 pc=0xc39fae
runtime.runfinq()
D:/opt/Go/src/runtime/mfinal.go:194 +0x107 fp=0xc00005ffe0 sp=0xc00005fe20 pc=0xc18347
runtime.goexit({})
D:/opt/Go/src/runtime/asm_amd64.s:1695 +0x1 fp=0xc00005ffe8 sp=0xc00005ffe0 pc=0xc65781
created by runtime.createfing in goroutine 1
D:/opt/Go/src/runtime/mfinal.go:164 +0x3d
rax 0x1fd7653258c
rbx 0x1fd76543ddc
rcx 0x1fd1e460000
rdx 0x1fd1e460000
rdi 0x386a0
rsi 0x1fd76549810
rbp 0x4e069ff480
rsp 0x4e069ff380
r8 0x1fd9553f020
r9 0x1
r10 0x1fd1e460000
r11 0x4e069ff250
r12 0x4e069ff630
r13 0x0
r14 0x1fd95e5c000
r15 0x3c31471d
rip 0x7ff91c7a3520
rflags 0x10293
cs 0x33
fs 0x53
gs 0x2b
我在阿里云的服务器上测试了Linux环境，由于没有音频设备，报了下面的错误
2024/04/06 12:52:21.578184 Failed to get default input device: %v
no default input device
不过我在阿里云的服务器上测试了non-streaming-decode-files，这个直接进行音频的语音识别的，所以运行正常
[root@iZ2zeatc350vkqmd1l53tjZ non-streaming-decode-files]# ./run-paraformer.sh
2024/04/06 12:56:21.554292 Reading ./sherpa-onnx-paraformer-trilingual-zh-cantonese-en/test_wavs/3-sichuan.wav
2024/04/06 12:56:21.571568 Initializing recognizer (may take several seconds)
2024/04/06 12:56:26.031247 Recognizer created!
2024/04/06 12:56:26.031276 Start decoding!
/project/sherpa-onnx/csrc/offline-paraformer-greedy-search-decoder.cc:Decode:65 time stamp for batch: 0, 40 vs -1
2024/04/06 12:56:26.626338 Decoding done!
2024/04/06 12:56:26.626369 自己就是在那个在那个就是在情节里面就是感觉是演得特别好就是好像很真实一样你知道吧
2024/04/06 12:56:26.626392 Wave duration: 7.835 seconds
今天又重新编译了，可以运行成功~
go version go1.22.2 windows/amd64
D:\a\sherpa-onnx\sherpa-onnx\sherpa-onnx\c-api\c-api.cc:SherpaOnnxCreateVoiceActivityDetector:812 VadModelConfig(silero_vad=SilerVadModelConfig(model="./silero_vad.onnx", threshold=0.5, min_silence_duration=0.5, min_speech_duration=0.25, window_size=512), sample_rate=16000, num_threads=1, provider="cpu", debug=True)
D:\a\sherpa-onnx\sherpa-onnx\sherpa-onnx\c-api\c-api.cc:CreateOfflineRecognizer:398 OfflineRecognizerConfig(feat_config=FeatureExtractorConfig(sampling_rate=16000, feature_dim=80, low_freq=20, high_freq=-400, dither=0), model_config=OfflineModelConfig(transducer=OfflineTransducerModelConfig(encoder_filename="", decoder_filename="", joiner_filename=""), paraformer=OfflineParaformerModelConfig(model="./sherpa-onnx-paraformer-trilingual-zh-cantonese-en/model.int8.onnx"), nemo_ctc=OfflineNemoEncDecCtcModelConfig(model=""), whisper=OfflineWhisperModelConfig(encoder="", decoder="", language="", task="transcribe", tail_paddings=-1), tdnn=OfflineTdnnModelConfig(model=""), zipformer_ctc=OfflineZipformerCtcModelConfig(model=""), wenet_ctc=OfflineWenetCtcModelConfig(model=""), telespeech_ctc="", tokens="./sherpa-onnx-paraformer-trilingual-zh-cantonese-en/tokens.txt", num_threads=2, debug=True, provider="cpu", model_type="", modeling_unit="", bpe_vocab=""), lm_config=OfflineLMConfig(model="", scale=1), ctc_fst_decoder_config=OfflineCtcFstDecoderConfig(graph="", max_active=3000), decoding_method="greedy_search", max_active_paths=4, hotwords_file="", hotwords_score=1.5, blank_penalty=0, rule_fsts="", rule_fars="")
2024/07/02 15:42:21.264935 Selected default input device: 麦克风阵列 (Realtek High Definition
2024/07/02 15:42:21.347389 Started! Please speak
2024/07/02 15:42:52.563583 Detected speech
2024/07/02 15:42:54.272830 Duration: 1.51 seconds
2024/07/02 15:42:54.348905 今天天气不错
2024/07/02 15:42:54.349189 Saved to seg-0-1.51-seconds-今天天气不错.wav
2024/07/02 15:42:54.349711 ----------