add alsa example for vad+offline asr (#2020)

cmake/cmake_extension.py

@@ -71,10 +71,11 @@ def get_binaries():
         binaries += [
             "sherpa-onnx-alsa",
             "sherpa-onnx-alsa-offline",
+            "sherpa-onnx-alsa-offline-audio-tagging",
             "sherpa-onnx-alsa-offline-speaker-identification",
             "sherpa-onnx-offline-tts-play-alsa",
             "sherpa-onnx-vad-alsa",
-            "sherpa-onnx-alsa-offline-audio-tagging",
+            "sherpa-onnx-vad-alsa-offline-asr",
         ]
 
     if is_windows():

sherpa-onnx/csrc/CMakeLists.txt

@@ -380,6 +380,7 @@ if(SHERPA_ONNX_HAS_ALSA AND SHERPA_ONNX_ENABLE_BINARY)
   add_executable(sherpa-onnx-alsa-offline-speaker-identification sherpa-onnx-alsa-offline-speaker-identification.cc alsa.cc)
   add_executable(sherpa-onnx-keyword-spotter-alsa sherpa-onnx-keyword-spotter-alsa.cc alsa.cc)
   add_executable(sherpa-onnx-vad-alsa sherpa-onnx-vad-alsa.cc alsa.cc)
+  add_executable(sherpa-onnx-vad-alsa-offline-asr sherpa-onnx-vad-alsa-offline-asr.cc alsa.cc)
 
 
   if(SHERPA_ONNX_ENABLE_TTS)
@@ -392,6 +393,7 @@ if(SHERPA_ONNX_HAS_ALSA AND SHERPA_ONNX_ENABLE_BINARY)
     sherpa-onnx-alsa-offline-speaker-identification
     sherpa-onnx-keyword-spotter-alsa
     sherpa-onnx-vad-alsa
+    sherpa-onnx-vad-alsa-offline-asr
     sherpa-onnx-alsa-offline-audio-tagging
   )
 

sherpa-onnx/csrc/sherpa-onnx-vad-alsa-offline-asr.cc

@@ -0,0 +1,159 @@
+// sherpa-onnx/csrc/sherpa-onnx-vad-alsa-offline-asr.cc
+//
+// Copyright (c)  2022-2025  Xiaomi Corporation
+
+#include <signal.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+#include <algorithm>
+#include <mutex>  // NOLINT
+
+#include "sherpa-onnx/csrc/alsa.h"
+#include "sherpa-onnx/csrc/circular-buffer.h"
+#include "sherpa-onnx/csrc/offline-recognizer.h"
+#include "sherpa-onnx/csrc/resample.h"
+#include "sherpa-onnx/csrc/voice-activity-detector.h"
+
+bool stop = false;
+static void Handler(int32_t /*sig*/) {
+  stop = true;
+  fprintf(stderr, "\nCaught Ctrl + C. Exiting...\n");
+}
+
+int32_t main(int32_t argc, char *argv[]) {
+  signal(SIGINT, Handler);
+
+  const char *kUsageMessage = R"usage(
+This program shows how to use a streaming VAD with non-streaming ASR in
+sherpa-onnx.
+
+Please download silero_vad.onnx from
+https://github.com/snakers4/silero-vad/raw/master/src/silero_vad/data/silero_vad.onnx
+
+For instance, use
+wget https://github.com/snakers4/silero-vad/raw/master/src/silero_vad/data/silero_vad.onnx
+
+Please refer to ./sherpa-onnx-microphone-offline.cc
+to download models for offline ASR.
+
+(1) Transducer from icefall
+
+  ./bin/sherpa-onnx-vad-microphone-offline-asr \
+    --silero-vad-model=/path/to/silero_vad.onnx \
+    --tokens=/path/to/tokens.txt \
+    --encoder=/path/to/encoder.onnx \
+    --decoder=/path/to/decoder.onnx \
+    --joiner=/path/to/joiner.onnx \
+    device_name
+
+(2) Paraformer from FunASR
+
+  ./bin/sherpa-onnx-vad-microphone-offline-asr \
+    --silero-vad-model=/path/to/silero_vad.onnx \
+    --tokens=/path/to/tokens.txt \
+    --paraformer=/path/to/model.onnx \
+    device_name
+
+(3) Whisper models
+
+  ./bin/sherpa-onnx-vad-microphone-offline-asr \
+    --silero-vad-model=/path/to/silero_vad.onnx \
+    --whisper-encoder=./sherpa-onnx-whisper-base.en/base.en-encoder.int8.onnx \
+    --whisper-decoder=./sherpa-onnx-whisper-base.en/base.en-decoder.int8.onnx \
+    --tokens=./sherpa-onnx-whisper-base.en/base.en-tokens.txt \
+    device_name
+
+The device name specifies which microphone to use in case there are several
+on your system. You can use
+
+  arecord -l
+
+to find all available microphones on your computer. For instance, if it outputs
+
+**** List of CAPTURE Hardware Devices ****
+card 3: UACDemoV10 [UACDemoV1.0], device 0: USB Audio [USB Audio]
+  Subdevices: 1/1
+  Subdevice #0: subdevice #0
+
+and if you want to select card 3 and device 0 on that card, please use:
+
+  plughw:3,0
+
+as the device_name.
+)usage";
+
+  sherpa_onnx::ParseOptions po(kUsageMessage);
+  sherpa_onnx::VadModelConfig vad_config;
+
+  sherpa_onnx::OfflineRecognizerConfig asr_config;
+
+  vad_config.Register(&po);
+  asr_config.Register(&po);
+
+  po.Read(argc, argv);
+  if (po.NumArgs() != 1) {
+    fprintf(stderr, "Please provide only 1 argument: the device name\n");
+    po.PrintUsage();
+    exit(EXIT_FAILURE);
+  }
+
+  fprintf(stderr, "%s\n", vad_config.ToString().c_str());
+  fprintf(stderr, "%s\n", asr_config.ToString().c_str());
+
+  if (!vad_config.Validate()) {
+    fprintf(stderr, "Errors in vad_config!\n");
+    return -1;
+  }
+
+  if (!asr_config.Validate()) {
+    fprintf(stderr, "Errors in asr_config!\n");
+    return -1;
+  }
+
+  fprintf(stderr, "Creating recognizer ...\n");
+  sherpa_onnx::OfflineRecognizer recognizer(asr_config);
+  fprintf(stderr, "Recognizer created!\n");
+
+  auto vad = std::make_unique<sherpa_onnx::VoiceActivityDetector>(vad_config);
+
+  std::string device_name = po.GetArg(1);
+  sherpa_onnx::Alsa alsa(device_name.c_str());
+  fprintf(stderr, "Use recording device: %s\n", device_name.c_str());
+
+  int32_t sample_rate = 16000;
+
+  if (alsa.GetExpectedSampleRate() != sample_rate) {
+    fprintf(stderr, "sample rate: %d != %d\n", alsa.GetExpectedSampleRate(),
+            sample_rate);
+    exit(-1);
+  }
+
+  int32_t chunk = 0.1 * alsa.GetActualSampleRate();
+
+  fprintf(stderr, "Started. Please speak\n");
+
+  int32_t window_size = vad_config.silero_vad.window_size;
+  int32_t index = 0;
+
+  while (!stop) {
+    const std::vector<float> &samples = alsa.Read(chunk);
+    vad->AcceptWaveform(samples.data(), samples.size());
+
+    while (!vad->Empty()) {
+      const auto &segment = vad->Front();
+      auto s = recognizer.CreateStream();
+      s->AcceptWaveform(sample_rate, segment.samples.data(),
+                        segment.samples.size());
+      recognizer.DecodeStream(s.get());
+      const auto &result = s->GetResult();
+      if (!result.text.empty()) {
+        fprintf(stderr, "%2d: %s\n", index, result.text.c_str());
+        ++index;
+      }
+      vad->Pop();
+    }
+  }
+
+  return 0;
+}

sherpa-onnx/csrc/sherpa-onnx-vad-microphone-offline-asr.cc

@@ -115,11 +115,20 @@ to download models for offline ASR.
 
   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();
 
   if (device_index == paNoDevice) {
     fprintf(stderr, "No default input device found\n");
+    fprintf(stderr,
+            "  If you are using Linux, please try "
+            "./build/bin/sherpa-onnx-vad-alsa-offline-asr\n");
     exit(EXIT_FAILURE);
   }
 

sherpa-onnx/csrc/vocoder.cc

@@ -87,6 +87,8 @@ std::unique_ptr<Vocoder> Vocoder::Create(const OfflineTtsModelConfig &config) {
       SHERPA_ONNX_LOGE("Unknown model type in vocoder!");
       return nullptr;
   }
+
+  return nullptr;
 }
 
 template <typename Manager>