Add silence threshold for alsa stream

common/message/wire_chunk.hpp

@@ -77,6 +77,12 @@ public:
     uint32_t payloadSize;
     char* payload;
 
+    template <typename T>
+    std::pair<T*, size_t> getPayload() const
+    {
+        return std::make_pair<T*, size_t>(reinterpret_cast<T*>(payload), payloadSize / sizeof(T));
+    }
+
 protected:
     void doserialize(std::ostream& stream) const override
     {

doc/configuration.md

@@ -142,13 +142,14 @@ output = audioresample ! audioconvert ! audio/x-raw,rate=48000,channels=2,format
 Captures audio from an alsa device
 
 ```sh
-alsa://?name=<name>&device=<alsa device>[&send_silence=false][&idle_threshold=100]
+alsa://?name=<name>&device=<alsa device>[&send_silence=false][&idle_threshold=100][&silence_threshold_percent=0.0]
 ```
 
 #### Available parameters
 
 - `device`: alsa device name or identifier, e.g. `default` or `hw:0,0` or `hw:0,0,0`
 - `idle_threshold`: switch stream state from playing to idle after receiving `idle_threshold` milliseconds of silence
+- `silence_threshold_percent`: percent (float) of the max amplitude to be considered as silence 
 - `send_silence`: forward silence to clients when stream state is `idle`
 
 The output of any audio player that uses alsa can be redirected to Snapcast by using an alsa loopback device:

server/etc/snapserver.conf

@@ -125,7 +125,7 @@ doc_root = /usr/share/snapserver/snapweb
 #  sampleformat will be set to "44100:16:2"
 # tcp server: tcp://<listen IP, e.g. 127.0.0.1>:<port>?name=<name>[&mode=server]
 # tcp client: tcp://<server IP, e.g. 127.0.0.1>:<port>?name=<name>&mode=client
-# alsa: alsa://?name=<name>&device=<alsa device>[&send_silence=false][&idle_threshold=100]
+# alsa: alsa://?name=<name>&device=<alsa device>[&send_silence=false][&idle_threshold=100][&silence_threshold_percent=0.0]
 # meta: meta:///<name of source#1>/<name of source#2>/.../<name of source#N>?name=<name>
 source = pipe:///tmp/snapfifo?name=default
 #source = tcp://127.0.0.1?name=mopidy_tcp

server/streamreader/alsa_stream.cpp

@@ -71,7 +71,18 @@ AlsaStream::AlsaStream(PcmListener* pcmListener, boost::asio::io_context& ioc, c
     device_ = uri_.getQuery("device", "hw:0");
     send_silence_ = (uri_.getQuery("send_silence", "false") == "true");
     idle_threshold_ = std::chrono::milliseconds(std::max(cpt::stoi(uri_.getQuery("idle_threshold", "100")), 10));
-    LOG(DEBUG, LOG_TAG) << "Device: " << device_ << "\n";
+    double silence_threshold_percent = 0.;
+    try
+    {
+        silence_threshold_percent = cpt::stod(uri_.getQuery("silence_threshold_percent", "0"));
+    }
+    catch (...)
+    {
+    }
+    int32_t max_amplitude = std::pow(2, sampleFormat_.bits() - 1) - 1;
+    silence_threshold_ = max_amplitude * (silence_threshold_percent / 100.);
+    LOG(DEBUG, LOG_TAG) << "Device: " << device_ << ", silence threshold percent: " << silence_threshold_percent
+                        << ", silence threshold amplitude: " << silence_threshold_ << "\n";
 }
 
 
@@ -166,6 +177,41 @@ void AlsaStream::uninitAlsa()
 }
 
 
+bool AlsaStream::isSilent(const msg::PcmChunk& chunk) const
+{
+    if (silence_threshold_ == 0)
+        return (std::memcmp(chunk.payload, silent_chunk_.data(), silent_chunk_.size()) == 0);
+
+    if (sampleFormat_.sampleSize() == 1)
+    {
+        auto payload = chunk.getPayload<int8_t>();
+        for (size_t n = 0; n < payload.second; ++n)
+        {
+            if (abs(payload.first[n]) > silence_threshold_)
+                return false;
+        }
+    }
+    else if (sampleFormat_.sampleSize() == 2)
+    {
+        auto payload = chunk.getPayload<int16_t>();
+        for (size_t n = 0; n < payload.second; ++n)
+        {
+            if (abs(payload.first[n]) > silence_threshold_)
+                return false;
+        }
+    }
+    else if (sampleFormat_.sampleSize() == 4)
+    {
+        auto payload = chunk.getPayload<int32_t>();
+        for (size_t n = 0; n < payload.second; ++n)
+        {
+            if (abs(payload.first[n]) > silence_threshold_)
+                return false;
+        }
+    }
+    return true;
+}
+
 void AlsaStream::do_read()
 {
     try
@@ -210,7 +256,7 @@ void AlsaStream::do_read()
             }
         } while (len < toRead);
 
-        if (std::memcmp(chunk_->payload, silent_chunk_.data(), silent_chunk_.size()) == 0)
+        if (isSilent(*chunk_))
         {
             silence_ += chunk_->duration<std::chrono::microseconds>();
             if (silence_ > idle_threshold_)

server/streamreader/alsa_stream.hpp

@@ -47,6 +47,9 @@ protected:
     void initAlsa();
     void uninitAlsa();
 
+    /// check if the chunk's volume is below the silence threshold
+    bool isSilent(const msg::PcmChunk& chunk) const;
+
     snd_pcm_t* handle_;
     std::unique_ptr<msg::PcmChunk> chunk_;
     bool first_;
@@ -61,6 +64,7 @@ protected:
     bool send_silence_;
     /// silence duration before switching the stream to idle
     std::chrono::milliseconds idle_threshold_;
+    int32_t silence_threshold_ = 0;
 };
 
 } // namespace streamreader