EngineX-Iluvatar/enginex-mr_series-sherpa-onnx
Archived
16
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

move portaudio common record code to microphone (#2264)

Browse Source 
Co-authored-by: cqm <cqm@97kid.com>
This commit is contained in:
mtdxc
2025-05-31 21:48:41 +08:00
committed by GitHub
parent 921f0f40cb
commit 613e8084c2
12 changed files with 130 additions and 446 deletions
Download Patch File Download Diff File Expand all files Collapse all files

55
cxx-api-examples/parakeet-tdt-simulate-streaming-microphone-cxx-api.cc
View File

@@ -116,7 +116,6 @@ int32_t main() {
sherpa_onnx::Microphone mic; sherpa_onnx::Microphone mic;
PaDeviceIndex num_devices = Pa_GetDeviceCount(); PaDeviceIndex num_devices = Pa_GetDeviceCount();
std::cout << "Num devices: " << num_devices << "\n";
if (num_devices == 0) { if (num_devices == 0) {
std::cerr << " If you are using Linux, please try " std::cerr << " If you are using Linux, please try "
"./build/bin/sense-voice-simulate-streaming-alsa-cxx-api\n"; "./build/bin/sense-voice-simulate-streaming-alsa-cxx-api\n";
@@ -124,39 +123,24 @@ int32_t main() {
} }
int32_t device_index = Pa_GetDefaultInputDevice(); int32_t device_index = Pa_GetDefaultInputDevice();
const char *pDeviceIndex = std::getenv("SHERPA_ONNX_MIC_DEVICE"); const char *pDeviceIndex = std::getenv("SHERPA_ONNX_MIC_DEVICE");
if (pDeviceIndex) { if (pDeviceIndex) {
fprintf(stderr, "Use specified device: %s\n", pDeviceIndex); fprintf(stderr, "Use specified device: %s\n", pDeviceIndex);
device_index = atoi(pDeviceIndex); device_index = atoi(pDeviceIndex);
} }
mic.PrintDevices(device_index);
for (int32_t i = 0; i != num_devices; ++i) {
const PaDeviceInfo *info = Pa_GetDeviceInfo(i);
fprintf(stderr, " %s %d %s\n", (i == device_index) ? "*" : " ", i,
info->name);
}
PaStreamParameters param;
param.device = device_index;
fprintf(stderr, "Use device: %d\n", param.device);
const PaDeviceInfo *info = Pa_GetDeviceInfo(param.device);
fprintf(stderr, " Name: %s\n", info->name);
fprintf(stderr, " Max input channels: %d\n", info->maxInputChannels);
param.channelCount = 1;
param.sampleFormat = paFloat32;
param.suggestedLatency = info->defaultLowInputLatency;
param.hostApiSpecificStreamInfo = nullptr;
float mic_sample_rate = 16000; float mic_sample_rate = 16000;
const char *sample_rate_str = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE"); const char *sample_rate_str = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE");
if (sample_rate_str) { if (sample_rate_str) {
fprintf(stderr, "Use sample rate %f for mic\n", mic_sample_rate); fprintf(stderr, "Use sample rate %f for mic\n", mic_sample_rate);
mic_sample_rate = atof(sample_rate_str); mic_sample_rate = atof(sample_rate_str);
} }
if(!mic.OpenDevice(device_index, mic_sample_rate, 1, RecordCallback,
nullptr) == false) {
std::cerr << "Failed to open microphone device\n";
return -1;
}
float sample_rate = 16000; float sample_rate = 16000;
LinearResampler resampler; LinearResampler resampler;
if (mic_sample_rate != sample_rate) { if (mic_sample_rate != sample_rate) {
@@ -168,27 +152,6 @@ int32_t main() {
lowpass_cutoff, lowpass_filter_width); lowpass_cutoff, lowpass_filter_width);
} }
PaStream *stream;
PaError err =
Pa_OpenStream(&stream, &param, nullptr, /* &outputParameters, */
mic_sample_rate,
0, // frames per buffer
paClipOff, // we won't output out of range samples
// so don't bother clipping them
RecordCallback, // RecordCallback is run in a separate
// thread created by portaudio
nullptr);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
err = Pa_StartStream(stream);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
int32_t window_size = 512; // samples, please don't change int32_t window_size = 512; // samples, please don't change
int32_t offset = 0; int32_t offset = 0;
@@ -276,11 +239,5 @@ int32_t main() {
} }
} }
err = Pa_CloseStream(stream);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
return 0; return 0;
} }

53
cxx-api-examples/sense-voice-simulate-streaming-microphone-cxx-api.cc
View File

@@ -112,7 +112,6 @@ int32_t main() {
sherpa_onnx::Microphone mic; sherpa_onnx::Microphone mic;
PaDeviceIndex num_devices = Pa_GetDeviceCount(); PaDeviceIndex num_devices = Pa_GetDeviceCount();
std::cout << "Num devices: " << num_devices << "\n";
if (num_devices == 0) { if (num_devices == 0) {
std::cerr << " If you are using Linux, please try " std::cerr << " If you are using Linux, please try "
"./build/bin/sense-voice-simulate-streaming-alsa-cxx-api\n"; "./build/bin/sense-voice-simulate-streaming-alsa-cxx-api\n";
@@ -120,33 +119,13 @@ int32_t main() {
} }
int32_t device_index = Pa_GetDefaultInputDevice(); int32_t device_index = Pa_GetDefaultInputDevice();
const char *pDeviceIndex = std::getenv("SHERPA_ONNX_MIC_DEVICE"); const char *pDeviceIndex = std::getenv("SHERPA_ONNX_MIC_DEVICE");
if (pDeviceIndex) { if (pDeviceIndex) {
fprintf(stderr, "Use specified device: %s\n", pDeviceIndex); fprintf(stderr, "Use specified device: %s\n", pDeviceIndex);
device_index = atoi(pDeviceIndex); device_index = atoi(pDeviceIndex);
} }
mic.PrintDevices(device_index);
for (int32_t i = 0; i != num_devices; ++i) {
const PaDeviceInfo *info = Pa_GetDeviceInfo(i);
fprintf(stderr, " %s %d %s\n", (i == device_index) ? "*" : " ", i,
info->name);
}
PaStreamParameters param;
param.device = device_index;
fprintf(stderr, "Use device: %d\n", param.device);
const PaDeviceInfo *info = Pa_GetDeviceInfo(param.device);
fprintf(stderr, " Name: %s\n", info->name);
fprintf(stderr, " Max input channels: %d\n", info->maxInputChannels);
param.channelCount = 1;
param.sampleFormat = paFloat32;
param.suggestedLatency = info->defaultLowInputLatency;
param.hostApiSpecificStreamInfo = nullptr;
float mic_sample_rate = 16000; float mic_sample_rate = 16000;
const char *sample_rate_str = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE"); const char *sample_rate_str = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE");
if (sample_rate_str) { if (sample_rate_str) {
@@ -163,26 +142,10 @@ int32_t main() {
resampler = LinearResampler::Create(mic_sample_rate, sample_rate, resampler = LinearResampler::Create(mic_sample_rate, sample_rate,
lowpass_cutoff, lowpass_filter_width); lowpass_cutoff, lowpass_filter_width);
} }
if (mic.OpenDevice(device_index, mic_sample_rate, 1, RecordCallback,
PaStream *stream; nullptr) == false) {
PaError err = std::cerr << "Failed to open microphone device\n";
Pa_OpenStream(&stream, &param, nullptr, /* &outputParameters, */ return -1;
mic_sample_rate,
0, // frames per buffer
paClipOff, // we won't output out of range samples
// so don't bother clipping them
RecordCallback, // RecordCallback is run in a separate
// thread created by portaudio
nullptr);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
err = Pa_StartStream(stream);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
} }
int32_t window_size = 512; // samples, please don't change int32_t window_size = 512; // samples, please don't change
@@ -272,11 +235,5 @@ int32_t main() {
} }
} }
err = Pa_CloseStream(stream);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
return 0; return 0;
} }

79
sherpa-onnx/csrc/microphone.cc
View File

@@ -7,8 +7,6 @@
#include <stdio.h> #include <stdio.h>
#include <stdlib.h> #include <stdlib.h>
#include "portaudio.h" // NOLINT
namespace sherpa_onnx { namespace sherpa_onnx {
Microphone::Microphone() { Microphone::Microphone() {
@@ -20,10 +18,85 @@ Microphone::Microphone() {
} }
Microphone::~Microphone() { Microphone::~Microphone() {
CloseDevice();
PaError err = Pa_Terminate(); PaError err = Pa_Terminate();
if (err != paNoError) { if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err)); fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(-1); }
}
int Microphone::GetDeviceCount() const {
return Pa_GetDeviceCount();
}
int Microphone::GetDefaultInputDevice() const {
return Pa_GetDefaultInputDevice();
}
void Microphone::PrintDevices(int device_index) const {
PaDeviceIndex num_devices = Pa_GetDeviceCount();
fprintf(stderr, "Num devices: %d\n", num_devices);
for (int32_t i = 0; i != num_devices; ++i) {
const PaDeviceInfo *info = Pa_GetDeviceInfo(i);
fprintf(stderr, " %s %d %s\n", (i == device_index) ? "*" : " ", i,
info->name);
}
}
bool Microphone::OpenDevice(int index, int sample_rate, int channel, PaStreamCallback cb, void* userdata) {
if (index < 0 || index >= Pa_GetDeviceCount()) {
fprintf(stderr, "Invalid device index: %d\n", index);
return false;
}
const PaDeviceInfo *info = Pa_GetDeviceInfo(index);
if (!info) {
fprintf(stderr, "No device info found for index: %d\n", index);
return false;
}
CloseDevice();
fprintf(stderr, "Use device: %d\n", index);
fprintf(stderr, " Name: %s\n", info->name);
fprintf(stderr, " Max input channels: %d\n", info->maxInputChannels);
PaStreamParameters param;
param.device = index;
param.channelCount = channel;
param.sampleFormat = paFloat32;
param.suggestedLatency = info->defaultLowInputLatency;
param.hostApiSpecificStreamInfo = nullptr;
PaError err = Pa_OpenStream(&stream, &param, nullptr, /* &outputParameters, */
sample_rate,
0, // frames per buffer
paClipOff, // we won't output out of range samples
// so don't bother clipping them
cb, userdata);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
return false;
}
err = Pa_StartStream(stream);
fprintf(stderr, "Started\n");
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
CloseDevice();
return false;
}
return true;
}
void Microphone::CloseDevice() {
if (stream) {
PaError err = Pa_CloseStream(stream);
if (err != paNoError) {
fprintf(stderr, "Pa_CloseStream error: %s\n", Pa_GetErrorText(err));
}
stream = nullptr;
} }
} }

9
sherpa-onnx/csrc/microphone.h
View File

@@ -4,13 +4,22 @@
#ifndef SHERPA_ONNX_CSRC_MICROPHONE_H_ #ifndef SHERPA_ONNX_CSRC_MICROPHONE_H_
#define SHERPA_ONNX_CSRC_MICROPHONE_H_ #define SHERPA_ONNX_CSRC_MICROPHONE_H_
#include "portaudio.h" // NOLINT
namespace sherpa_onnx { namespace sherpa_onnx {
class Microphone { class Microphone {
PaStream *stream = nullptr;
public: public:
Microphone(); Microphone();
~Microphone(); ~Microphone();
int GetDeviceCount() const;
int GetDefaultInputDevice() const;
void PrintDevices(int sel) const;
bool OpenDevice(int index, int sample_rate, int channel, PaStreamCallback cb, void* userdata);
void CloseDevice();
}; };
} // namespace sherpa_onnx } // namespace sherpa_onnx

1
sherpa-onnx/csrc/sherpa-onnx-alsa-offline-speaker-identification.cc
View File

@@ -14,7 +14,6 @@
#include "sherpa-onnx/csrc/alsa.h" #include "sherpa-onnx/csrc/alsa.h"
#include "sherpa-onnx/csrc/macros.h" #include "sherpa-onnx/csrc/macros.h"
#include "sherpa-onnx/csrc/microphone.h"
#include "sherpa-onnx/csrc/speaker-embedding-extractor.h" #include "sherpa-onnx/csrc/speaker-embedding-extractor.h"
#include "sherpa-onnx/csrc/speaker-embedding-manager.h" #include "sherpa-onnx/csrc/speaker-embedding-manager.h"
#include "sherpa-onnx/csrc/wave-reader.h" #include "sherpa-onnx/csrc/wave-reader.h"

52
sherpa-onnx/csrc/sherpa-onnx-keyword-spotter-microphone.cc
View File

@@ -79,11 +79,7 @@ for a list of pre-trained models to download.
sherpa_onnx::Microphone mic; sherpa_onnx::Microphone mic;
PaDeviceIndex num_devices = Pa_GetDeviceCount();
fprintf(stderr, "Num devices: %d\n", num_devices);
int32_t device_index = Pa_GetDefaultInputDevice(); int32_t device_index = Pa_GetDefaultInputDevice();
if (device_index == paNoDevice) { if (device_index == paNoDevice) {
fprintf(stderr, "No default input device found\n"); fprintf(stderr, "No default input device found\n");
fprintf(stderr, "If you are using Linux, please switch to \n"); fprintf(stderr, "If you are using Linux, please switch to \n");
@@ -97,26 +93,7 @@ for a list of pre-trained models to download.
device_index = atoi(pDeviceIndex); device_index = atoi(pDeviceIndex);
} }
for (int32_t i = 0; i != num_devices; ++i) { mic.PrintDevices(device_index);
const PaDeviceInfo *info = Pa_GetDeviceInfo(i);
fprintf(stderr, " %s %d %s\n", (i == device_index) ? "*" : " ", i,
info->name);
}
PaStreamParameters param;
param.device = device_index;
fprintf(stderr, "Use device: %d\n", param.device);
const PaDeviceInfo *info = Pa_GetDeviceInfo(param.device);
fprintf(stderr, " Name: %s\n", info->name);
fprintf(stderr, " Max input channels: %d\n", info->maxInputChannels);
param.channelCount = 1;
param.sampleFormat = paFloat32;
param.suggestedLatency = info->defaultLowInputLatency;
param.hostApiSpecificStreamInfo = nullptr;
const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE"); const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE");
if (pSampleRateStr) { if (pSampleRateStr) {
@@ -124,24 +101,9 @@ for a list of pre-trained models to download.
mic_sample_rate = atof(pSampleRateStr); mic_sample_rate = atof(pSampleRateStr);
} }
PaStream *stream; if(!mic.OpenDevice(device_index, mic_sample_rate, 1,
PaError err = RecordCallback, s.get())) {
Pa_OpenStream(&stream, &param, nullptr, /* &outputParameters, */ fprintf(stderr, "portaudio error: %d\n", device_index);
mic_sample_rate,
0, // frames per buffer
paClipOff, // we won't output out of range samples
// so don't bother clipping them
RecordCallback, s.get());
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
err = Pa_StartStream(stream);
fprintf(stderr, "Started\n");
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE); exit(EXIT_FAILURE);
} }
@@ -164,11 +126,5 @@ for a list of pre-trained models to download.
Pa_Sleep(20); // sleep for 20ms Pa_Sleep(20); // sleep for 20ms
} }
err = Pa_CloseStream(stream);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
return 0; return 0;
} }

52
sherpa-onnx/csrc/sherpa-onnx-microphone-offline-audio-tagging.cc
View File

@@ -120,11 +120,7 @@ for more models.
sherpa_onnx::Microphone mic; sherpa_onnx::Microphone mic;
PaDeviceIndex num_devices = Pa_GetDeviceCount();
fprintf(stderr, "Num devices: %d\n", num_devices);
int32_t device_index = Pa_GetDefaultInputDevice(); int32_t device_index = Pa_GetDefaultInputDevice();
if (device_index == paNoDevice) { if (device_index == paNoDevice) {
fprintf(stderr, "No default input device found\n"); fprintf(stderr, "No default input device found\n");
fprintf(stderr, "If you are using Linux, please switch to \n"); fprintf(stderr, "If you are using Linux, please switch to \n");
@@ -138,26 +134,7 @@ for more models.
device_index = atoi(pDeviceIndex); device_index = atoi(pDeviceIndex);
} }
for (int32_t i = 0; i != num_devices; ++i) { mic.PrintDevices(device_index);
const PaDeviceInfo *info = Pa_GetDeviceInfo(i);
fprintf(stderr, " %s %d %s\n", (i == device_index) ? "*" : " ", i,
info->name);
}
PaStreamParameters param;
param.device = device_index;
fprintf(stderr, "Use device: %d\n", param.device);
const PaDeviceInfo *info = Pa_GetDeviceInfo(param.device);
fprintf(stderr, " Name: %s\n", info->name);
fprintf(stderr, " Max input channels: %d\n", info->maxInputChannels);
param.channelCount = 1;
param.sampleFormat = paFloat32;
param.suggestedLatency = info->defaultLowInputLatency;
param.hostApiSpecificStreamInfo = nullptr;
float mic_sample_rate = 16000; float mic_sample_rate = 16000;
const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE"); const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE");
if (pSampleRateStr) { if (pSampleRateStr) {
@@ -165,24 +142,9 @@ for more models.
mic_sample_rate = atof(pSampleRateStr); mic_sample_rate = atof(pSampleRateStr);
} }
PaStream *stream; if (!mic.OpenDevice(device_index, mic_sample_rate, 1,
PaError err = RecordCallback, nullptr /* user_data */)){
Pa_OpenStream(&stream, &param, nullptr, /* &outputParameters, */ fprintf(stderr, "portaudio error: %d\n", device_index);
mic_sample_rate,
0, // frames per buffer
paClipOff, // we won't output out of range samples
// so don't bother clipping them
RecordCallback, nullptr);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
err = Pa_StartStream(stream);
fprintf(stderr, "Started\n");
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE); exit(EXIT_FAILURE);
} }
@@ -226,11 +188,5 @@ for more models.
} }
t.join(); t.join();
err = Pa_CloseStream(stream);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
return 0; return 0;
} }

51
sherpa-onnx/csrc/sherpa-onnx-microphone-offline-speaker-identification.cc
View File

@@ -220,9 +220,6 @@ Note that `zh` means Chinese, while `en` means English.
sherpa_onnx::Microphone mic; sherpa_onnx::Microphone mic;
PaDeviceIndex num_devices = Pa_GetDeviceCount();
fprintf(stderr, "Num devices: %d\n", num_devices);
int32_t device_index = Pa_GetDefaultInputDevice(); int32_t device_index = Pa_GetDefaultInputDevice();
if (device_index == paNoDevice) { if (device_index == paNoDevice) {
fprintf(stderr, "No default input device found\n"); fprintf(stderr, "No default input device found\n");
@@ -238,52 +235,18 @@ Note that `zh` means Chinese, while `en` means English.
device_index = atoi(pDeviceIndex); device_index = atoi(pDeviceIndex);
} }
for (int32_t i = 0; i != num_devices; ++i) { mic.PrintDevices(device_index);
const PaDeviceInfo *info = Pa_GetDeviceInfo(i);
fprintf(stderr, " %s %d %s\n", (i == device_index) ? "*" : " ", i,
info->name);
}
PaStreamParameters param;
param.device = device_index;
fprintf(stderr, "Use device: %d\n", param.device);
const PaDeviceInfo *info = Pa_GetDeviceInfo(param.device);
fprintf(stderr, " Name: %s\n", info->name);
fprintf(stderr, " Max input channels: %d\n", info->maxInputChannels);
param.channelCount = 1;
param.sampleFormat = paFloat32;
param.suggestedLatency = info->defaultLowInputLatency;
param.hostApiSpecificStreamInfo = nullptr;
float mic_sample_rate = 16000; float mic_sample_rate = 16000;
const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE"); const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE");
if (pSampleRateStr) { if (pSampleRateStr) {
fprintf(stderr, "Use sample rate %f for mic\n", mic_sample_rate); fprintf(stderr, "Use sample rate %f for mic\n", mic_sample_rate);
mic_sample_rate = atof(pSampleRateStr); mic_sample_rate = atof(pSampleRateStr);
} }
float sample_rate = 16000;
PaStream *stream; if (!mic.OpenDevice(device_index, mic_sample_rate, 1,
PaError err = RecordCallback, nullptr /* user_data */)){
Pa_OpenStream(&stream, &param, nullptr, /* &outputParameters, */ fprintf(stderr, "portaudio error: %d\n", device_index);
mic_sample_rate,
0, // frames per buffer
paClipOff, // we won't output out of range samples
// so don't bother clipping them
RecordCallback, nullptr);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
err = Pa_StartStream(stream);
fprintf(stderr, "Started\n");
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE); exit(EXIT_FAILURE);
} }
@@ -323,11 +286,5 @@ Note that `zh` means Chinese, while `en` means English.
} }
t.join(); t.join();
err = Pa_CloseStream(stream);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
return 0; return 0;
} }

51
sherpa-onnx/csrc/sherpa-onnx-microphone-offline.cc
View File

@@ -136,11 +136,7 @@ for a list of pre-trained models to download.
sherpa_onnx::Microphone mic; sherpa_onnx::Microphone mic;
PaDeviceIndex num_devices = Pa_GetDeviceCount();
fprintf(stderr, "Num devices: %d\n", num_devices);
int32_t device_index = Pa_GetDefaultInputDevice(); int32_t device_index = Pa_GetDefaultInputDevice();
if (device_index == paNoDevice) { if (device_index == paNoDevice) {
fprintf(stderr, "No default input device found\n"); fprintf(stderr, "No default input device found\n");
fprintf(stderr, "If you are using Linux, please switch to \n"); fprintf(stderr, "If you are using Linux, please switch to \n");
@@ -154,26 +150,8 @@ for a list of pre-trained models to download.
device_index = atoi(pDeviceIndex); device_index = atoi(pDeviceIndex);
} }
for (int32_t i = 0; i != num_devices; ++i) { mic.PrintDevices(device_index);
const PaDeviceInfo *info = Pa_GetDeviceInfo(i);
fprintf(stderr, " %s %d %s\n", (i == device_index) ? "*" : " ", i,
info->name);
}
PaStreamParameters param;
param.device = device_index;
fprintf(stderr, "Use device: %d\n", param.device);
const PaDeviceInfo *info = Pa_GetDeviceInfo(param.device);
fprintf(stderr, " Name: %s\n", info->name);
fprintf(stderr, " Max input channels: %d\n", info->maxInputChannels);
param.channelCount = 1;
param.sampleFormat = paFloat32;
param.suggestedLatency = info->defaultLowInputLatency;
param.hostApiSpecificStreamInfo = nullptr;
float mic_sample_rate = 16000; float mic_sample_rate = 16000;
const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE"); const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE");
if (pSampleRateStr) { if (pSampleRateStr) {
@@ -181,24 +159,9 @@ for a list of pre-trained models to download.
mic_sample_rate = atof(pSampleRateStr); mic_sample_rate = atof(pSampleRateStr);
} }
PaStream *stream; if (!mic.OpenDevice(device_index, mic_sample_rate, 1,
PaError err = RecordCallback, nullptr /* user_data */)){
Pa_OpenStream(&stream, &param, nullptr, /* &outputParameters, */ fprintf(stderr, "portaudio error: %d\n", device_index);
mic_sample_rate,
0, // frames per buffer
paClipOff, // we won't output out of range samples
// so don't bother clipping them
RecordCallback, nullptr);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
err = Pa_StartStream(stream);
fprintf(stderr, "Started\n");
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE); exit(EXIT_FAILURE);
} }
@@ -232,11 +195,5 @@ for a list of pre-trained models to download.
} }
t.join(); t.join();
err = Pa_CloseStream(stream);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
return 0; return 0;
} }

53
sherpa-onnx/csrc/sherpa-onnx-microphone.cc
View File

@@ -106,11 +106,7 @@ for a list of pre-trained models to download.
sherpa_onnx::Microphone mic; sherpa_onnx::Microphone mic;
PaDeviceIndex num_devices = Pa_GetDeviceCount();
fprintf(stderr, "Num devices: %d\n", num_devices);
int32_t device_index = Pa_GetDefaultInputDevice(); int32_t device_index = Pa_GetDefaultInputDevice();
if (device_index == paNoDevice) { if (device_index == paNoDevice) {
fprintf(stderr, "No default input device found\n"); fprintf(stderr, "No default input device found\n");
fprintf(stderr, "If you are using Linux, please switch to \n"); fprintf(stderr, "If you are using Linux, please switch to \n");
@@ -124,51 +120,18 @@ for a list of pre-trained models to download.
device_index = atoi(pDeviceIndex); device_index = atoi(pDeviceIndex);
} }
for (int32_t i = 0; i != num_devices; ++i) { mic.PrintDevices(device_index);
const PaDeviceInfo *info = Pa_GetDeviceInfo(i);
fprintf(stderr, " %s %d %s\n", (i == device_index) ? "*" : " ", i,
info->name);
}
PaStreamParameters param; float mic_sample_rate = 16000;
param.device = device_index;
fprintf(stderr, "Use device: %d\n", param.device);
const PaDeviceInfo *info = Pa_GetDeviceInfo(param.device);
fprintf(stderr, " Name: %s\n", info->name);
fprintf(stderr, " Max input channels: %d\n", info->maxInputChannels);
param.channelCount = 1;
param.sampleFormat = paFloat32;
param.suggestedLatency = info->defaultLowInputLatency;
param.hostApiSpecificStreamInfo = nullptr;
const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE"); const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE");
if (pSampleRateStr) { if (pSampleRateStr) {
fprintf(stderr, "Use sample rate %f for mic\n", mic_sample_rate); fprintf(stderr, "Use sample rate %f for mic\n", mic_sample_rate);
mic_sample_rate = atof(pSampleRateStr); mic_sample_rate = atof(pSampleRateStr);
} }
float sample_rate = 16000;
PaStream *stream; if (!mic.OpenDevice(device_index, mic_sample_rate, 1,
PaError err = RecordCallback, nullptr /* user_data */)){
Pa_OpenStream(&stream, &param, nullptr, /* &outputParameters, */ fprintf(stderr, "portaudio error: %d\n", device_index);
sample_rate,
0, // frames per buffer
paClipOff, // we won't output out of range samples
// so don't bother clipping them
RecordCallback, s.get());
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
err = Pa_StartStream(stream);
fprintf(stderr, "Started\n");
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE); exit(EXIT_FAILURE);
} }
@@ -213,11 +176,5 @@ for a list of pre-trained models to download.
Pa_Sleep(20); // sleep for 20ms Pa_Sleep(20); // sleep for 20ms
} }
err = Pa_CloseStream(stream);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
return 0; return 0;
} }

62
sherpa-onnx/csrc/sherpa-onnx-vad-microphone-offline-asr.cc
View File

@@ -113,17 +113,7 @@ to download models for offline ASR.
sherpa_onnx::Microphone mic; sherpa_onnx::Microphone mic;
PaDeviceIndex num_devices = Pa_GetDeviceCount();
fprintf(stderr, "Num devices: %d\n", num_devices);
if (num_devices == 0) {
fprintf(stderr,
" If you are using Linux, please try "
"./build/bin/sherpa-onnx-vad-alsa-offline-asr\n");
exit(-1);
}
int32_t device_index = Pa_GetDefaultInputDevice(); int32_t device_index = Pa_GetDefaultInputDevice();
if (device_index == paNoDevice) { if (device_index == paNoDevice) {
fprintf(stderr, "No default input device found\n"); fprintf(stderr, "No default input device found\n");
fprintf(stderr, fprintf(stderr,
@@ -137,33 +127,20 @@ to download models for offline ASR.
fprintf(stderr, "Use specified device: %s\n", pDeviceIndex); fprintf(stderr, "Use specified device: %s\n", pDeviceIndex);
device_index = atoi(pDeviceIndex); device_index = atoi(pDeviceIndex);
} }
mic.PrintDevices(device_index);
for (int32_t i = 0; i != num_devices; ++i) {
const PaDeviceInfo *info = Pa_GetDeviceInfo(i);
fprintf(stderr, " %s %d %s\n", (i == device_index) ? "*" : " ", i,
info->name);
}
PaStreamParameters param;
param.device = device_index;
fprintf(stderr, "Use device: %d\n", param.device);
const PaDeviceInfo *info = Pa_GetDeviceInfo(param.device);
fprintf(stderr, " Name: %s\n", info->name);
fprintf(stderr, " Max input channels: %d\n", info->maxInputChannels);
param.channelCount = 1;
param.sampleFormat = paFloat32;
param.suggestedLatency = info->defaultLowInputLatency;
param.hostApiSpecificStreamInfo = nullptr;
float mic_sample_rate = 16000; float mic_sample_rate = 16000;
const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE"); const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE");
if (pSampleRateStr) { if (pSampleRateStr) {
fprintf(stderr, "Use sample rate %f for mic\n", mic_sample_rate); fprintf(stderr, "Use sample rate %f for mic\n", mic_sample_rate);
mic_sample_rate = atof(pSampleRateStr); mic_sample_rate = atof(pSampleRateStr);
} }
if (!mic.OpenDevice(device_index, mic_sample_rate, 1, RecordCallback, nullptr)) {
fprintf(stderr, "Failed to open device %d\n", device_index);
exit(EXIT_FAILURE);
}
float sample_rate = 16000; float sample_rate = 16000;
std::unique_ptr<sherpa_onnx::LinearResample> resampler; std::unique_ptr<sherpa_onnx::LinearResample> resampler;
if (mic_sample_rate != sample_rate) { if (mic_sample_rate != sample_rate) {
@@ -175,25 +152,6 @@ to download models for offline ASR.
mic_sample_rate, sample_rate, lowpass_cutoff, lowpass_filter_width); mic_sample_rate, sample_rate, lowpass_cutoff, lowpass_filter_width);
} }
PaStream *stream;
PaError err =
Pa_OpenStream(&stream, &param, nullptr, /* &outputParameters, */
mic_sample_rate,
0, // frames per buffer
paClipOff, // we won't output out of range samples
// so don't bother clipping them
RecordCallback, nullptr);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
err = Pa_StartStream(stream);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
auto vad = std::make_unique<sherpa_onnx::VoiceActivityDetector>(vad_config); auto vad = std::make_unique<sherpa_onnx::VoiceActivityDetector>(vad_config);
fprintf(stderr, "Started. Please speak\n"); fprintf(stderr, "Started. Please speak\n");
@@ -236,11 +194,5 @@ to download models for offline ASR.
Pa_Sleep(100); // sleep for 100ms Pa_Sleep(100); // sleep for 100ms
} }
err = Pa_CloseStream(stream);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
return 0; return 0;
} }

58
sherpa-onnx/csrc/sherpa-onnx-vad-microphone.cc
View File

@@ -74,11 +74,8 @@ wget https://github.com/snakers4/silero-vad/raw/master/src/silero_vad/data/siler
sherpa_onnx::Microphone mic; sherpa_onnx::Microphone mic;
PaDeviceIndex num_devices = Pa_GetDeviceCount();
fprintf(stderr, "Num devices: %d\n", num_devices);
int32_t device_index = Pa_GetDefaultInputDevice(); int32_t device_index = Pa_GetDefaultInputDevice();
if (device_index == paNoDevice) { if (device_index == paNoDevice) {
fprintf(stderr, "No default input device found\n"); fprintf(stderr, "No default input device found\n");
fprintf(stderr, "If you are using Linux, please switch to \n"); fprintf(stderr, "If you are using Linux, please switch to \n");
@@ -91,35 +88,20 @@ wget https://github.com/snakers4/silero-vad/raw/master/src/silero_vad/data/siler
fprintf(stderr, "Use specified device: %s\n", pDeviceIndex); fprintf(stderr, "Use specified device: %s\n", pDeviceIndex);
device_index = atoi(pDeviceIndex); device_index = atoi(pDeviceIndex);
} }
mic.PrintDevices(device_index);
for (int32_t i = 0; i != num_devices; ++i) {
const PaDeviceInfo *info = Pa_GetDeviceInfo(i);
fprintf(stderr, " %s %d %s\n", (i == device_index) ? "*" : " ", i,
info->name);
}
PaStreamParameters param;
param.device = device_index;
fprintf(stderr, "Use device: %d\n", param.device);
const PaDeviceInfo *info = Pa_GetDeviceInfo(param.device);
fprintf(stderr, " Name: %s\n", info->name);
fprintf(stderr, " Max input channels: %d\n", info->maxInputChannels);
param.channelCount = 1;
param.sampleFormat = paFloat32;
param.suggestedLatency = info->defaultLowInputLatency;
param.hostApiSpecificStreamInfo = nullptr;
float mic_sample_rate = 16000; float mic_sample_rate = 16000;
const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE"); const char *pSampleRateStr = std::getenv("SHERPA_ONNX_MIC_SAMPLE_RATE");
if (pSampleRateStr) { if (pSampleRateStr) {
fprintf(stderr, "Use sample rate %f for mic\n", mic_sample_rate); fprintf(stderr, "Use sample rate %f for mic\n", mic_sample_rate);
mic_sample_rate = atof(pSampleRateStr); mic_sample_rate = atof(pSampleRateStr);
} }
float sample_rate = 16000; if (!mic.OpenDevice(device_index, mic_sample_rate, 1, RecordCallback, nullptr)) {
fprintf(stderr, "Failed to open microphone device %d\n", device_index);
exit(EXIT_FAILURE);
}
float sample_rate = 16000;
std::unique_ptr<sherpa_onnx::LinearResample> resampler; std::unique_ptr<sherpa_onnx::LinearResample> resampler;
if (mic_sample_rate != sample_rate) { if (mic_sample_rate != sample_rate) {
float min_freq = std::min(mic_sample_rate, sample_rate); float min_freq = std::min(mic_sample_rate, sample_rate);
@@ -130,30 +112,8 @@ wget https://github.com/snakers4/silero-vad/raw/master/src/silero_vad/data/siler
mic_sample_rate, sample_rate, lowpass_cutoff, lowpass_filter_width); mic_sample_rate, sample_rate, lowpass_cutoff, lowpass_filter_width);
} }
PaStream *stream;
PaError err =
Pa_OpenStream(&stream, &param, nullptr, /* &outputParameters, */
mic_sample_rate,
0, // frames per buffer
paClipOff, // we won't output out of range samples
// so don't bother clipping them
RecordCallback, nullptr);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
err = Pa_StartStream(stream);
auto vad = std::make_unique<sherpa_onnx::VoiceActivityDetector>(config); auto vad = std::make_unique<sherpa_onnx::VoiceActivityDetector>(config);
fprintf(stderr, "Started\n");
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
int32_t window_size = config.silero_vad.window_size; int32_t window_size = config.silero_vad.window_size;
bool printed = false; bool printed = false;
@@ -202,11 +162,5 @@ wget https://github.com/snakers4/silero-vad/raw/master/src/silero_vad/data/siler
Pa_Sleep(100); // sleep for 100ms Pa_Sleep(100); // sleep for 100ms
} }
err = Pa_CloseStream(stream);
if (err != paNoError) {
fprintf(stderr, "portaudio error: %s\n", Pa_GetErrorText(err));
exit(EXIT_FAILURE);
}
return 0; return 0;
} }