devttl/xiaozhi-esp32
1
0
Fork 0
forked from xiaozhi/xiaozhi-esp32
Code Pull Requests Activity

update protocol to support manual response mode

Browse Source
This commit is contained in:
Terrence
2024-11-25 00:59:03 +08:00
parent aa806f676e
commit 472219d5bf
11 changed files with 166 additions and 100 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
main/wake_word_detect.cc
View File

@@ -4,9 +4,9 @@
#include <esp_log.h>
#include <model_path.h>
#include <arpa/inet.h>
#include <sstream>
#define DETECTION_RUNNING_EVENT 1
#define WAKE_WORD_ENCODED_EVENT 2
static const char* TAG = "WakeWordDetect";
@@ -40,6 +40,13 @@ void WakeWordDetect::Initialize(int channels, bool reference) {
ESP_LOGI(TAG, "Model %d: %s", i, models->model_name[i]);
if (strstr(models->model_name[i], ESP_WN_PREFIX) != NULL) {
wakenet_model_ = models->model_name[i];
auto words = esp_srmodel_get_wake_words(models, wakenet_model_);
// split by ";" to get all wake words
std::stringstream ss(words);
std::string word;
while (std::getline(ss, word, ';')) {
wake_words_.push_back(word);
}
}
}
@@ -84,7 +91,7 @@ void WakeWordDetect::Initialize(int channels, bool reference) {
}, "audio_detection", 4096 * 2, this, 1, nullptr);
}
void WakeWordDetect::OnWakeWordDetected(std::function<void()> callback) {
void WakeWordDetect::OnWakeWordDetected(std::function<void(const std::string& wake_word)> callback) {
wake_word_detected_callback_ = callback;
}
@@ -144,11 +151,11 @@ void WakeWordDetect::AudioDetectionTask() {
}
if (res->wakeup_state == WAKENET_DETECTED) {
ESP_LOGI(TAG, "Wake word detected");
StopDetection();
last_detected_wake_word_ = wake_words_[res->wake_word_index - 1];
if (wake_word_detected_callback_) {
wake_word_detected_callback_();
wake_word_detected_callback_(last_detected_wake_word_);
}
}
}
@@ -165,7 +172,6 @@ void WakeWordDetect::StoreWakeWordData(uint16_t* data, size_t samples) {
}
void WakeWordDetect::EncodeWakeWordData() {
xEventGroupClearBits(event_group_, WAKE_WORD_ENCODED_EVENT);
wake_word_opus_.clear();
if (wake_word_encode_task_stack_ == nullptr) {
wake_word_encode_task_stack_ = (StackType_t*)heap_caps_malloc(4096 * 8, MALLOC_CAP_SPIRAM);
@@ -182,15 +188,18 @@ void WakeWordDetect::EncodeWakeWordData() {
encoder->Encode(pcm, [this_](const uint8_t* opus, size_t opus_size) {
std::lock_guard<std::mutex> lock(this_->wake_word_mutex_);
this_->wake_word_opus_.emplace_back(std::string(reinterpret_cast<const char*>(opus), opus_size));
this_->wake_word_cv_.notify_one();
this_->wake_word_cv_.notify_all();
});
}
this_->wake_word_pcm_.clear();
auto end_time = esp_timer_get_time();
ESP_LOGI(TAG, "Encode wake word opus %zu packets in %lld ms", this_->wake_word_opus_.size(), (end_time - start_time) / 1000);
xEventGroupSetBits(this_->event_group_, WAKE_WORD_ENCODED_EVENT);
this_->wake_word_cv_.notify_one();
{
std::lock_guard<std::mutex> lock(this_->wake_word_mutex_);
this_->wake_word_opus_.push_back("");
this_->wake_word_cv_.notify_all();
}
delete encoder;
vTaskDelete(NULL);
}, "encode_detect_packets", 4096 * 8, this, 1, wake_word_encode_task_stack_, &wake_word_encode_task_buffer_);
@@ -199,12 +208,9 @@ void WakeWordDetect::EncodeWakeWordData() {
bool WakeWordDetect::GetWakeWordOpus(std::string& opus) {
std::unique_lock<std::mutex> lock(wake_word_mutex_);
wake_word_cv_.wait(lock, [this]() {
return !wake_word_opus_.empty() || (xEventGroupGetBits(event_group_) & WAKE_WORD_ENCODED_EVENT);
return !wake_word_opus_.empty();
});
if (wake_word_opus_.empty()) {
return false;
}
opus.swap(wake_word_opus_.front());
wake_word_opus_.pop_front();
return true;
return !opus.empty();
}