Only reset VAD state between chunks of activity

locaal-ai · Aug 2, 2024 · 606938d · 606938d
1 parent 1ed62ca
commit 606938d
Show file tree

Hide file tree

Showing 3 changed files with 11 additions and 9 deletions.
diff --git a/src/whisper-utils/silero-vad-onnx.cpp b/src/whisper-utils/silero-vad-onnx.cpp
@@ -92,11 +92,13 @@ void VadIterator::init_onnx_model(const SileroString &model_path)
 	session = std::make_shared<Ort::Session>(env, model_path.c_str(), session_options);
 };
 
-void VadIterator::reset_states()
+void VadIterator::reset_states(bool reset_state)
 {
-	// Call reset before each audio start
-	std::memset(_state.data(), 0.0f, _state.size() * sizeof(float));
-	triggered = false;
+	if (reset_state) {
+		// Call reset before each audio start
+		std::memset(_state.data(), 0.0f, _state.size() * sizeof(float));
+		triggered = false;
+	}
 	temp_end = 0;
 	current_sample = 0;
 
@@ -257,9 +259,9 @@ void VadIterator::predict(const std::vector<float> &data)
 	}
 };
 
-void VadIterator::process(const std::vector<float> &input_wav)
+void VadIterator::process(const std::vector<float> &input_wav, bool reset_state)
 {
-	reset_states();
+	reset_states(reset_state);
 
 	audio_length_samples = (int)input_wav.size();
 

diff --git a/src/whisper-utils/silero-vad-onnx.h b/src/whisper-utils/silero-vad-onnx.h
@@ -43,12 +43,12 @@ class VadIterator {
 private:
 	void init_engine_threads(int inter_threads, int intra_threads);
 	void init_onnx_model(const SileroString &model_path);
-	void reset_states();
+	void reset_states(bool reset_state);
 	float predict_one(const std::vector<float> &data);
 	void predict(const std::vector<float> &data);
 
 public:
-	void process(const std::vector<float> &input_wav);
+	void process(const std::vector<float> &input_wav, bool reset_state = true);
 	void process(const std::vector<float> &input_wav, std::vector<float> &output_wav);
 	void collect_chunks(const std::vector<float> &input_wav, std::vector<float> &output_wav);
 	const std::vector<timestamp_t> get_speech_timestamps() const;

diff --git a/src/whisper-utils/whisper-processing.cpp b/src/whisper-utils/whisper-processing.cpp
@@ -426,7 +426,7 @@ vad_state vad_based_segmentation(transcription_filter_data *gf, vad_state last_v
 	obs_log(gf->log_level, "sending %d frames to vad", vad_input.size());
 	{
 		ProfileScope("vad->process");
-		gf->vad->process(vad_input);
+		gf->vad->process(vad_input, !last_vad_state.vad_on);
 	}
 
 	const uint64_t start_ts_offset_ms = start_timestamp_offset_ns / 1000000;