Voice Assistant improvements (#5827)

esphome/components/i2s_audio/speaker/i2s_audio_speaker.cpp

@@ -220,6 +220,8 @@ size_t I2SAudioSpeaker::play(const uint8_t *data, size_t length) {
   return index;
 }
 
+bool I2SAudioSpeaker::has_buffered_data() const { return uxQueueMessagesWaiting(this->buffer_queue_) > 0; }
+
 }  // namespace i2s_audio
 }  // namespace esphome
 

esphome/components/i2s_audio/speaker/i2s_audio_speaker.h

@@ -56,6 +56,8 @@ class I2SAudioSpeaker : public Component, public speaker::Speaker, public I2SAud
 
   size_t play(const uint8_t *data, size_t length) override;
 
+  bool has_buffered_data() const override;
+
  protected:
   void start_();
   // void stop_();

esphome/components/speaker/speaker.h

@@ -18,6 +18,8 @@ class Speaker {
   virtual void start() = 0;
   virtual void stop() = 0;
 
+  virtual bool has_buffered_data() const = 0;
+
   bool is_running() const { return this->state_ == STATE_RUNNING; }
 
  protected:

esphome/components/voice_assistant/voice_assistant.cpp

@@ -273,28 +273,27 @@ void VoiceAssistant::loop() {
       bool playing = false;
 #ifdef USE_SPEAKER
       if (this->speaker_ != nullptr) {
+        ssize_t received_len = 0;
         if (this->speaker_buffer_index_ + RECEIVE_SIZE < SPEAKER_BUFFER_SIZE) {
-          auto len = this->socket_->read(this->speaker_buffer_ + this->speaker_buffer_index_, RECEIVE_SIZE);
+          received_len = this->socket_->read(this->speaker_buffer_ + this->speaker_buffer_index_, RECEIVE_SIZE);
-          if (len > 0) {
+          if (received_len > 0) {
-            this->speaker_buffer_index_ += len;
+            this->speaker_buffer_index_ += received_len;
-            this->speaker_buffer_size_ += len;
+            this->speaker_buffer_size_ += received_len;
+            this->speaker_bytes_received_ += received_len;
           }
         } else {
-          ESP_LOGW(TAG, "Receive buffer full");
+          ESP_LOGD(TAG, "Receive buffer full");
-        }
-        if (this->speaker_buffer_size_ > 0) {
-          size_t written = this->speaker_->play(this->speaker_buffer_, this->speaker_buffer_size_);
-          if (written > 0) {
-            memmove(this->speaker_buffer_, this->speaker_buffer_ + written, this->speaker_buffer_size_ - written);
-            this->speaker_buffer_size_ -= written;
-            this->speaker_buffer_index_ -= written;
-            this->set_timeout("speaker-timeout", 2000, [this]() { this->speaker_->stop(); });
-          } else {
-            ESP_LOGW(TAG, "Speaker buffer full");
-          }
         }
+        // Build a small buffer of audio before sending to the speaker
+        if (this->speaker_bytes_received_ > RECEIVE_SIZE * 4)
+          this->write_speaker_();
         if (this->wait_for_stream_end_) {
           this->cancel_timeout("playing");
+          if (this->stream_ended_ && received_len < 0) {
+            ESP_LOGD(TAG, "End of audio stream received");
+            this->cancel_timeout("speaker-timeout");
+            this->set_state_(State::RESPONSE_FINISHED, State::RESPONSE_FINISHED);
+          }
           break;  // We dont want to timeout here as the STREAM_END event will take care of that.
         }
         playing = this->speaker_->is_running();
@@ -316,14 +315,26 @@ void VoiceAssistant::loop() {
     case State::RESPONSE_FINISHED: {
 #ifdef USE_SPEAKER
       if (this->speaker_ != nullptr) {
+        if (this->speaker_buffer_size_ > 0) {
+          this->write_speaker_();
+          break;
+        }
+        if (this->speaker_->has_buffered_data() || this->speaker_->is_running()) {
+          break;
+        }
+        ESP_LOGD(TAG, "Speaker has finished outputting all audio");
         this->speaker_->stop();
         this->cancel_timeout("speaker-timeout");
         this->cancel_timeout("playing");
         this->speaker_buffer_size_ = 0;
         this->speaker_buffer_index_ = 0;
+        this->speaker_bytes_received_ = 0;
         memset(this->speaker_buffer_, 0, SPEAKER_BUFFER_SIZE);
+        this->wait_for_stream_end_ = false;
+        this->stream_ended_ = false;
+
+        this->tts_stream_end_trigger_->trigger();
       }
-      this->wait_for_stream_end_ = false;
 #endif
       this->set_state_(State::IDLE, State::IDLE);
       break;
@@ -333,6 +344,20 @@ void VoiceAssistant::loop() {
   }
 }
 
+void VoiceAssistant::write_speaker_() {
+  if (this->speaker_buffer_size_ > 0) {
+    size_t written = this->speaker_->play(this->speaker_buffer_, this->speaker_buffer_size_);
+    if (written > 0) {
+      memmove(this->speaker_buffer_, this->speaker_buffer_ + written, this->speaker_buffer_size_ - written);
+      this->speaker_buffer_size_ -= written;
+      this->speaker_buffer_index_ -= written;
+      this->set_timeout("speaker-timeout", 5000, [this]() { this->speaker_->stop(); });
+    } else {
+      ESP_LOGD(TAG, "Speaker buffer full, trying again next loop");
+    }
+  }
+}
+
 void VoiceAssistant::client_subscription(api::APIConnection *client, bool subscribe) {
   if (!subscribe) {
     if (this->api_client_ == nullptr || client != this->api_client_) {
@@ -503,21 +528,20 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) {
   switch (msg.event_type) {
     case api::enums::VOICE_ASSISTANT_RUN_START:
       ESP_LOGD(TAG, "Assist Pipeline running");
-      this->start_trigger_->trigger();
+      this->defer([this]() { this->start_trigger_->trigger(); });
       break;
     case api::enums::VOICE_ASSISTANT_WAKE_WORD_START:
       break;
     case api::enums::VOICE_ASSISTANT_WAKE_WORD_END: {
       ESP_LOGD(TAG, "Wake word detected");
-      this->wake_word_detected_trigger_->trigger();
+      this->defer([this]() { this->wake_word_detected_trigger_->trigger(); });
       break;
     }
     case api::enums::VOICE_ASSISTANT_STT_START:
       ESP_LOGD(TAG, "STT started");
-      this->listening_trigger_->trigger();
+      this->defer([this]() { this->listening_trigger_->trigger(); });
       break;
     case api::enums::VOICE_ASSISTANT_STT_END: {
-      this->set_state_(State::STOP_MICROPHONE, State::AWAITING_RESPONSE);
       std::string text;
       for (auto arg : msg.data) {
         if (arg.name == "text") {
@@ -529,12 +553,12 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) {
         return;
       }
       ESP_LOGD(TAG, "Speech recognised as: \"%s\"", text.c_str());
-      this->stt_end_trigger_->trigger(text);
+      this->defer([this, text]() { this->stt_end_trigger_->trigger(text); });
       break;
     }
     case api::enums::VOICE_ASSISTANT_INTENT_START:
       ESP_LOGD(TAG, "Intent started");
-      this->intent_start_trigger_->trigger();
+      this->defer([this]() { this->intent_start_trigger_->trigger(); });
       break;
     case api::enums::VOICE_ASSISTANT_INTENT_END: {
       for (auto arg : msg.data) {
@@ -542,7 +566,7 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) {
           this->conversation_id_ = std::move(arg.value);
         }
       }
-      this->intent_end_trigger_->trigger();
+      this->defer([this]() { this->intent_end_trigger_->trigger(); });
       break;
     }
     case api::enums::VOICE_ASSISTANT_TTS_START: {
@@ -557,10 +581,12 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) {
         return;
       }
       ESP_LOGD(TAG, "Response: \"%s\"", text.c_str());
-      this->tts_start_trigger_->trigger(text);
+      this->defer([this, text]() {
+        this->tts_start_trigger_->trigger(text);
 #ifdef USE_SPEAKER
-      this->speaker_->start();
+        this->speaker_->start();
 #endif
+      });
       break;
     }
     case api::enums::VOICE_ASSISTANT_TTS_END: {
@@ -575,14 +601,16 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) {
         return;
       }
       ESP_LOGD(TAG, "Response URL: \"%s\"", url.c_str());
+      this->defer([this, url]() {
 #ifdef USE_MEDIA_PLAYER
-      if (this->media_player_ != nullptr) {
+        if (this->media_player_ != nullptr) {
-        this->media_player_->make_call().set_media_url(url).perform();
+          this->media_player_->make_call().set_media_url(url).perform();
-      }
+        }
 #endif
+        this->tts_end_trigger_->trigger(url);
+      });
       State new_state = this->local_output_ ? State::STREAMING_RESPONSE : State::IDLE;
       this->set_state_(new_state, new_state);
-      this->tts_end_trigger_->trigger(url);
       break;
     }
     case api::enums::VOICE_ASSISTANT_RUN_END: {
@@ -599,7 +627,7 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) {
           this->set_state_(State::IDLE, State::IDLE);
         }
       }
-      this->end_trigger_->trigger();
+      this->defer([this]() { this->end_trigger_->trigger(); });
       break;
     }
     case api::enums::VOICE_ASSISTANT_ERROR: {
@@ -617,8 +645,10 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) {
         return;
       } else if (code == "wake-provider-missing" || code == "wake-engine-missing") {
         // Wake word is not set up or not ready on Home Assistant so stop and do not retry until user starts again.
-        this->request_stop();
+        this->defer([this, code, message]() {
-        this->error_trigger_->trigger(code, message);
+          this->request_stop();
+          this->error_trigger_->trigger(code, message);
+        });
         return;
       }
       ESP_LOGE(TAG, "Error: %s - %s", code.c_str(), message.c_str());
@@ -626,32 +656,32 @@ void VoiceAssistant::on_event(const api::VoiceAssistantEventResponse &msg) {
         this->signal_stop_();
         this->set_state_(State::STOP_MICROPHONE, State::IDLE);
       }
-      this->error_trigger_->trigger(code, message);
+      this->defer([this, code, message]() { this->error_trigger_->trigger(code, message); });
       break;
     }
     case api::enums::VOICE_ASSISTANT_TTS_STREAM_START: {
 #ifdef USE_SPEAKER
       this->wait_for_stream_end_ = true;
       ESP_LOGD(TAG, "TTS stream start");
-      this->tts_stream_start_trigger_->trigger();
+      this->defer([this] { this->tts_stream_start_trigger_->trigger(); });
 #endif
       break;
     }
     case api::enums::VOICE_ASSISTANT_TTS_STREAM_END: {
-      this->set_state_(State::RESPONSE_FINISHED, State::IDLE);
 #ifdef USE_SPEAKER
+      this->stream_ended_ = true;
       ESP_LOGD(TAG, "TTS stream end");
-      this->tts_stream_end_trigger_->trigger();
 #endif
       break;
     }
     case api::enums::VOICE_ASSISTANT_STT_VAD_START:
       ESP_LOGD(TAG, "Starting STT by VAD");
-      this->stt_vad_start_trigger_->trigger();
+      this->defer([this]() { this->stt_vad_start_trigger_->trigger(); });
       break;
     case api::enums::VOICE_ASSISTANT_STT_VAD_END:
       ESP_LOGD(TAG, "STT by VAD end");
-      this->stt_vad_end_trigger_->trigger();
+      this->set_state_(State::STOP_MICROPHONE, State::AWAITING_RESPONSE);
+      this->defer([this]() { this->stt_vad_end_trigger_->trigger(); });
       break;
     default:
       ESP_LOGD(TAG, "Unhandled event type: %d", msg.event_type);

esphome/components/voice_assistant/voice_assistant.h

@@ -156,11 +156,14 @@ class VoiceAssistant : public Component {
 
   microphone::Microphone *mic_{nullptr};
 #ifdef USE_SPEAKER
+  void write_speaker_();
   speaker::Speaker *speaker_{nullptr};
   uint8_t *speaker_buffer_;
   size_t speaker_buffer_index_{0};
   size_t speaker_buffer_size_{0};
+  size_t speaker_bytes_received_{0};
   bool wait_for_stream_end_{false};
+  bool stream_ended_{false};
 #endif
 #ifdef USE_MEDIA_PLAYER
   media_player::MediaPlayer *media_player_{nullptr};