#include "custom_wake_word.h"
|
#include "audio_service.h"
|
#include "system_info.h"
|
#include "assets.h"
|
#include "boards/sening-xiaozhi-s3/system.h"
|
|
#include <esp_log.h>
|
#include <esp_mn_iface.h>
|
#include <esp_mn_models.h>
|
#include <esp_mn_speech_commands.h>
|
#include <cJSON.h>
|
#include <chrono>
|
|
#define TAG "CustomWakeWord"
|
|
CustomWakeWord::CustomWakeWord()
|
: wake_word_pcm_(), wake_word_opus_() {
|
}
|
|
CustomWakeWord::~CustomWakeWord() {
|
if (multinet_model_data_ != nullptr && multinet_ != nullptr) {
|
multinet_->destroy(multinet_model_data_);
|
multinet_model_data_ = nullptr;
|
}
|
|
if (wake_word_encode_task_stack_ != nullptr) {
|
heap_caps_free(wake_word_encode_task_stack_);
|
}
|
|
if (wake_word_encode_task_buffer_ != nullptr) {
|
heap_caps_free(wake_word_encode_task_buffer_);
|
}
|
|
if (models_ != nullptr) {
|
esp_srmodel_deinit(models_);
|
}
|
}
|
|
void CustomWakeWord::ParseWakenetModelConfig() {
|
// Read index.json
|
auto& assets = Assets::GetInstance();
|
void* ptr = nullptr;
|
size_t size = 0;
|
if (!assets.GetAssetData("index.json", ptr, size)) {
|
ESP_LOGE(TAG, "Failed to read index.json");
|
return;
|
}
|
cJSON* root = cJSON_ParseWithLength(static_cast<char*>(ptr), size);
|
if (root == nullptr) {
|
ESP_LOGE(TAG, "Failed to parse index.json");
|
return;
|
}
|
cJSON* multinet_model = cJSON_GetObjectItem(root, "multinet_model");
|
if (cJSON_IsObject(multinet_model)) {
|
cJSON* language = cJSON_GetObjectItem(multinet_model, "language");
|
cJSON* duration = cJSON_GetObjectItem(multinet_model, "duration");
|
cJSON* threshold = cJSON_GetObjectItem(multinet_model, "threshold");
|
cJSON* commands = cJSON_GetObjectItem(multinet_model, "commands");
|
if (cJSON_IsString(language)) {
|
language_ = language->valuestring;
|
}
|
if (cJSON_IsNumber(duration)) {
|
duration_ = duration->valueint;
|
}
|
if (cJSON_IsNumber(threshold)) {
|
threshold_ = threshold->valuedouble;
|
}
|
if (cJSON_IsArray(commands)) {
|
for (int i = 0; i < cJSON_GetArraySize(commands); i++) {
|
cJSON* command = cJSON_GetArrayItem(commands, i);
|
if (cJSON_IsObject(command)) {
|
cJSON* command_name = cJSON_GetObjectItem(command, "command");
|
cJSON* text = cJSON_GetObjectItem(command, "text");
|
cJSON* action = cJSON_GetObjectItem(command, "action");
|
if (cJSON_IsString(command_name) && cJSON_IsString(text) && cJSON_IsString(action)) {
|
commands_.push_back({command_name->valuestring, text->valuestring, action->valuestring});
|
ESP_LOGI(TAG, "Command: %s, Text: %s, Action: %s", command_name->valuestring, text->valuestring, action->valuestring);
|
}
|
}
|
}
|
}
|
}
|
cJSON_Delete(root);
|
}
|
|
|
bool CustomWakeWord::Initialize(AudioCodec* codec, srmodel_list_t* models_list) {
|
codec_ = codec;
|
commands_.clear();
|
|
if (models_list == nullptr) {
|
language_ = "cn";
|
models_ = esp_srmodel_init("model");
|
#if defined(BOARD_CUSTOM_WAKE_WORD_PINYIN) && defined(BOARD_CUSTOM_WAKE_WORD_DISPLAY) && defined(BOARD_CUSTOM_WAKE_WORD_THRESHOLD_PERCENT)
|
threshold_ = BOARD_CUSTOM_WAKE_WORD_THRESHOLD_PERCENT / 100.0f;
|
commands_.push_back({BOARD_CUSTOM_WAKE_WORD_PINYIN, BOARD_CUSTOM_WAKE_WORD_DISPLAY, "wake"});
|
#elif defined(CONFIG_CUSTOM_WAKE_WORD)
|
threshold_ = CONFIG_CUSTOM_WAKE_WORD_THRESHOLD / 100.0f;
|
commands_.push_back({CONFIG_CUSTOM_WAKE_WORD, CONFIG_CUSTOM_WAKE_WORD_DISPLAY, "wake"});
|
#endif
|
commands_.push_back({OFFLINE_COMMAND_WORD_PINYIN, OFFLINE_COMMAND_WORD, "offline_command"});
|
commands_.push_back({OFFLINE_COMMAND_WORD_ALT_PINYIN, OFFLINE_COMMAND_WORD_ALT, "offline_command"});
|
commands_.push_back({OFFLINE_REPEATER_WORD_PINYIN, OFFLINE_REPEATER_WORD, "offline_command"});
|
commands_.push_back({OFFLINE_START_REPEATER_WORD_PINYIN, OFFLINE_START_REPEATER_WORD, "offline_command"});
|
commands_.push_back({OFFLINE_EXIT_REPEATER_WORD_PINYIN, OFFLINE_EXIT_REPEATER_WORD, "offline_command"});
|
commands_.push_back({OFFLINE_EXIT_SHADOWING_WORD_PINYIN, OFFLINE_EXIT_SHADOWING_WORD, "offline_command"});
|
commands_.push_back({OFFLINE_CHAT_WORD_PINYIN, OFFLINE_CHAT_WORD, "offline_command"});
|
commands_.push_back({OFFLINE_CHAT_WORD_FULL_PINYIN, OFFLINE_CHAT_WORD_FULL, "offline_command"});
|
} else {
|
models_ = models_list;
|
ParseWakenetModelConfig();
|
#if defined(BOARD_CUSTOM_WAKE_WORD_PINYIN) && defined(BOARD_CUSTOM_WAKE_WORD_DISPLAY) && defined(BOARD_CUSTOM_WAKE_WORD_THRESHOLD_PERCENT)
|
threshold_ = BOARD_CUSTOM_WAKE_WORD_THRESHOLD_PERCENT / 100.0f;
|
for (auto it = commands_.begin(); it != commands_.end();) {
|
if (it->action == "wake") {
|
it = commands_.erase(it);
|
} else {
|
++it;
|
}
|
}
|
commands_.push_back({BOARD_CUSTOM_WAKE_WORD_PINYIN, BOARD_CUSTOM_WAKE_WORD_DISPLAY, "wake"});
|
#endif
|
commands_.push_back({OFFLINE_COMMAND_WORD_PINYIN, OFFLINE_COMMAND_WORD, "offline_command"});
|
commands_.push_back({OFFLINE_COMMAND_WORD_ALT_PINYIN, OFFLINE_COMMAND_WORD_ALT, "offline_command"});
|
commands_.push_back({OFFLINE_REPEATER_WORD_PINYIN, OFFLINE_REPEATER_WORD, "offline_command"});
|
commands_.push_back({OFFLINE_START_REPEATER_WORD_PINYIN, OFFLINE_START_REPEATER_WORD, "offline_command"});
|
commands_.push_back({OFFLINE_EXIT_REPEATER_WORD_PINYIN, OFFLINE_EXIT_REPEATER_WORD, "offline_command"});
|
commands_.push_back({OFFLINE_EXIT_SHADOWING_WORD_PINYIN, OFFLINE_EXIT_SHADOWING_WORD, "offline_command"});
|
commands_.push_back({OFFLINE_CHAT_WORD_PINYIN, OFFLINE_CHAT_WORD, "offline_command"});
|
commands_.push_back({OFFLINE_CHAT_WORD_FULL_PINYIN, OFFLINE_CHAT_WORD_FULL, "offline_command"});
|
}
|
|
if (models_ == nullptr || models_->num == -1) {
|
ESP_LOGE(TAG, "Failed to initialize wakenet model");
|
return false;
|
}
|
|
// 初始化 multinet (命令词识别)
|
mn_name_ = esp_srmodel_filter(models_, ESP_MN_PREFIX, language_.c_str());
|
if (mn_name_ == nullptr) {
|
ESP_LOGW(TAG, "Language '%s' multinet not found, falling back to any multinet model", language_.c_str());
|
mn_name_ = esp_srmodel_filter(models_, ESP_MN_PREFIX, NULL);
|
}
|
if (mn_name_ == nullptr) {
|
ESP_LOGE(TAG, "Failed to initialize multinet, mn_name is nullptr");
|
ESP_LOGI(TAG, "Please refer to https://pcn7cs20v8cr.feishu.cn/wiki/CpQjwQsCJiQSWSkYEvrcxcbVnwh to add custom wake word");
|
return false;
|
}
|
|
multinet_ = esp_mn_handle_from_name(mn_name_);
|
multinet_model_data_ = multinet_->create(mn_name_, duration_);
|
multinet_->set_det_threshold(multinet_model_data_, threshold_);
|
esp_mn_commands_clear();
|
for (int i = 0; i < commands_.size(); i++) {
|
esp_mn_commands_add(i + 1, commands_[i].command.c_str());
|
}
|
esp_mn_commands_update();
|
|
multinet_->print_active_speech_commands(multinet_model_data_);
|
return true;
|
}
|
|
void CustomWakeWord::OnWakeWordDetected(std::function<void(const std::string& wake_word)> callback) {
|
wake_word_detected_callback_ = callback;
|
}
|
|
void CustomWakeWord::OnOfflineCommandDetected(std::function<void(const std::string& command)> callback) {
|
offline_command_detected_callback_ = callback;
|
}
|
|
void CustomWakeWord::Start() {
|
running_ = true;
|
}
|
|
void CustomWakeWord::Stop() {
|
running_ = false;
|
|
std::lock_guard<std::mutex> lock(input_buffer_mutex_);
|
input_buffer_.clear();
|
}
|
|
void CustomWakeWord::Feed(const std::vector<int16_t>& data) {
|
if (multinet_model_data_ == nullptr) {
|
return;
|
}
|
|
std::lock_guard<std::mutex> lock(input_buffer_mutex_);
|
// Check running state inside lock to avoid TOCTOU race with Stop()
|
if (!running_) {
|
return;
|
}
|
|
// If input channels is 2, we need to fetch the left channel data
|
if (codec_->input_channels() == 2) {
|
for (size_t i = 0; i < data.size(); i += 2) {
|
input_buffer_.push_back(data[i]);
|
}
|
} else {
|
input_buffer_.insert(input_buffer_.end(), data.begin(), data.end());
|
}
|
|
int chunksize = multinet_->get_samp_chunksize(multinet_model_data_);
|
while (input_buffer_.size() >= chunksize) {
|
std::vector<int16_t> chunk(input_buffer_.begin(), input_buffer_.begin() + chunksize);
|
StoreWakeWordData(chunk);
|
|
esp_mn_state_t mn_state = multinet_->detect(multinet_model_data_, chunk.data());
|
|
if (mn_state == ESP_MN_STATE_DETECTED) {
|
esp_mn_results_t *mn_result = multinet_->get_results(multinet_model_data_);
|
for (int i = 0; i < mn_result->num && running_; i++) {
|
int command_id = mn_result->command_id[i];
|
ESP_LOGI(TAG, "Custom wake word detected: command_id=%d, string=%s, prob=%f",
|
command_id, mn_result->string, mn_result->prob[i]);
|
if (command_id < 1 || command_id > static_cast<int>(commands_.size())) {
|
continue;
|
}
|
auto& command = commands_[command_id - 1];
|
if (command.action == "wake") {
|
last_detected_wake_word_ = command.text;
|
running_ = false;
|
input_buffer_.clear();
|
|
if (wake_word_detected_callback_) {
|
wake_word_detected_callback_(last_detected_wake_word_);
|
}
|
} else if (command.action == "offline_command") {
|
if (offline_command_detected_callback_) {
|
offline_command_detected_callback_(command.text);
|
}
|
}
|
}
|
multinet_->clean(multinet_model_data_);
|
} else if (mn_state == ESP_MN_STATE_TIMEOUT) {
|
ESP_LOGD(TAG, "Command word detection timeout, cleaning state");
|
multinet_->clean(multinet_model_data_);
|
}
|
|
if (!running_) {
|
break;
|
}
|
input_buffer_.erase(input_buffer_.begin(), input_buffer_.begin() + chunksize);
|
}
|
}
|
|
size_t CustomWakeWord::GetFeedSize() {
|
if (multinet_model_data_ == nullptr) {
|
return 0;
|
}
|
return multinet_->get_samp_chunksize(multinet_model_data_);
|
}
|
|
void CustomWakeWord::StoreWakeWordData(const std::vector<int16_t>& data) {
|
std::vector<int16_t> chunk;
|
chunk.reserve(2048);
|
chunk.assign(data.begin(), data.end());
|
wake_word_pcm_.push_back(std::move(chunk));
|
// keep about 2 seconds of data, detect duration is 30ms (sample_rate == 16000, chunksize == 512)
|
while (wake_word_pcm_.size() > 2000 / 30) {
|
wake_word_pcm_.pop_front();
|
}
|
}
|
|
void CustomWakeWord::ClearWakeBuffers() {
|
std::lock_guard<std::mutex> lock(wake_word_mutex_);
|
wake_word_pcm_.clear();
|
wake_word_opus_.clear();
|
}
|
|
void CustomWakeWord::AbortWakeWordEncode() {
|
std::lock_guard<std::mutex> lock(wake_word_mutex_);
|
wake_word_pcm_.clear();
|
wake_word_opus_.clear();
|
wake_word_opus_.emplace_back();
|
wake_word_cv_.notify_all();
|
}
|
|
void CustomWakeWord::EncodeWakeWordData() {
|
const size_t stack_size = 4096 * 7;
|
wake_word_opus_.clear();
|
if (wake_word_encode_task_stack_ == nullptr) {
|
wake_word_encode_task_stack_ = (StackType_t*)heap_caps_malloc(stack_size, MALLOC_CAP_SPIRAM);
|
assert(wake_word_encode_task_stack_ != nullptr);
|
}
|
if (wake_word_encode_task_buffer_ == nullptr) {
|
wake_word_encode_task_buffer_ = (StaticTask_t*)heap_caps_malloc(sizeof(StaticTask_t), MALLOC_CAP_INTERNAL);
|
assert(wake_word_encode_task_buffer_ != nullptr);
|
}
|
|
wake_word_encode_task_ = xTaskCreateStatic([](void* arg) {
|
auto this_ = (CustomWakeWord*)arg;
|
{
|
auto start_time = esp_timer_get_time();
|
// Create encoder
|
esp_opus_enc_config_t opus_enc_cfg = AS_OPUS_ENC_CONFIG();
|
void* encoder_handle = nullptr;
|
auto ret = esp_opus_enc_open(&opus_enc_cfg, sizeof(esp_opus_enc_config_t), &encoder_handle);
|
if (encoder_handle == nullptr) {
|
ESP_LOGE(TAG, "Failed to create audio encoder, error code: %d", ret);
|
std::lock_guard<std::mutex> lock(this_->wake_word_mutex_);
|
this_->wake_word_opus_.push_back(std::vector<uint8_t>());
|
this_->wake_word_cv_.notify_all();
|
return;
|
}
|
// Get frame size
|
int frame_size = 0;
|
int outbuf_size = 0;
|
esp_opus_enc_get_frame_size(encoder_handle, &frame_size, &outbuf_size);
|
frame_size = frame_size / sizeof(int16_t);
|
// Encode all PCM data
|
int packets = 0;
|
std::vector<int16_t> in_buffer;
|
esp_audio_enc_in_frame_t in = {};
|
esp_audio_enc_out_frame_t out = {};
|
for (auto& pcm: this_->wake_word_pcm_) {
|
if (in_buffer.empty()) {
|
in_buffer = std::move(pcm);
|
} else {
|
in_buffer.reserve(in_buffer.size() + pcm.size());
|
in_buffer.insert(in_buffer.end(), pcm.begin(), pcm.end());
|
}
|
while (in_buffer.size() >= frame_size) {
|
std::vector<uint8_t> opus_buf(outbuf_size);
|
in.buffer = (uint8_t *)(in_buffer.data());
|
in.len = (uint32_t)(frame_size * sizeof(int16_t));
|
out.buffer = opus_buf.data();
|
out.len = outbuf_size;
|
out.encoded_bytes = 0;
|
ret = esp_opus_enc_process(encoder_handle, &in, &out);
|
if (ret == ESP_AUDIO_ERR_OK) {
|
std::lock_guard<std::mutex> lock(this_->wake_word_mutex_);
|
this_->wake_word_opus_.emplace_back(opus_buf.data(), opus_buf.data() + out.encoded_bytes);
|
this_->wake_word_cv_.notify_all();
|
packets++;
|
} else {
|
ESP_LOGE(TAG, "Failed to encode audio, error code: %d", ret);
|
}
|
in_buffer.erase(in_buffer.begin(), in_buffer.begin() + frame_size);
|
}
|
}
|
this_->wake_word_pcm_.clear();
|
// Close encoder
|
esp_opus_enc_close(encoder_handle);
|
auto end_time = esp_timer_get_time();
|
ESP_LOGI(TAG, "Encode wake word opus %d packets in %ld ms", packets, (long)((end_time - start_time) / 1000));
|
|
std::lock_guard<std::mutex> lock(this_->wake_word_mutex_);
|
this_->wake_word_opus_.push_back(std::vector<uint8_t>());
|
this_->wake_word_cv_.notify_all();
|
}
|
vTaskDelete(NULL);
|
}, "encode_wake_word", stack_size, this, 2, wake_word_encode_task_stack_, wake_word_encode_task_buffer_);
|
if (wake_word_encode_task_ == nullptr) {
|
ESP_LOGE(TAG, "Failed to create encode_wake_word task");
|
std::lock_guard<std::mutex> lock(wake_word_mutex_);
|
wake_word_opus_.push_back(std::vector<uint8_t>());
|
wake_word_cv_.notify_all();
|
}
|
}
|
|
bool CustomWakeWord::GetWakeWordOpus(std::vector<uint8_t>& opus) {
|
std::unique_lock<std::mutex> lock(wake_word_mutex_);
|
if (!wake_word_cv_.wait_for(lock, std::chrono::milliseconds(2000), [this]() {
|
return !wake_word_opus_.empty();
|
})) {
|
ESP_LOGW(TAG, "wake opus wait timeout");
|
return false;
|
}
|
opus.swap(wake_word_opus_.front());
|
wake_word_opus_.pop_front();
|
return !opus.empty();
|
}
|
