kanshan/xiaozhi-esp32
1
0
Fork 0
mirror of https://github.com/78/xiaozhi-esp32.git synced 2026-01-14 01:07:30 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Add SetInputGain(float gain) to AudioCodec (#1252)
Some checks are pending
Build Boards / Determine variants to build (push) Waiting to run
Details
Build Boards / Build ${{ matrix.name }} (push) Blocked by required conditions
Details

Browse Source
This commit is contained in:
Xiaoxia 2025-10-02 09:55:45 +08:00 committed by GitHub
parent 911fee2d85
commit 2fd0ff0b2a
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
10 changed files with 80 additions and 115 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
main/audio/audio_codec.cc
View File

@ -55,6 +55,11 @@ void AudioCodec::SetOutputVolume(int volume) {
settings.SetInt("output_volume", output_volume_);
}
void AudioCodec::SetInputGain(float gain) {
input_gain_ = gain;
ESP_LOGI(TAG, "Set input gain to %.1f", input_gain_);
}
void AudioCodec::EnableInput(bool enable) {
if (enable == input_enabled_) {
return;

4
main/audio/audio_codec.h
View File

@ -13,7 +13,6 @@
#define AUDIO_CODEC_DMA_DESC_NUM 6
#define AUDIO_CODEC_DMA_FRAME_NUM 240
#define AUDIO_CODEC_DEFAULT_MIC_GAIN 30.0
class AudioCodec {
public:
@ -21,6 +20,7 @@ public:
virtual ~AudioCodec();
virtual void SetOutputVolume(int volume);
virtual void SetInputGain(float gain);
virtual void EnableInput(bool enable);
virtual void EnableOutput(bool enable);
@ -35,6 +35,7 @@ public:
inline int input_channels() const { return input_channels_; }
inline int output_channels() const { return output_channels_; }
inline int output_volume() const { return output_volume_; }
inline float input_gain() const { return input_gain_; }
inline bool input_enabled() const { return input_enabled_; }
inline bool output_enabled() const { return output_enabled_; }
@ -51,6 +52,7 @@ protected:
int input_channels_ = 1;
int output_channels_ = 1;
int output_volume_ = 70;
float input_gain_ = 0.0;
virtual int Read(int16_t* dest, int samples) = 0;
virtual int Write(const int16_t* data, int samples) = 0;

3
main/audio/codecs/box_audio_codec.cc
View File

@ -14,6 +14,7 @@ BoxAudioCodec::BoxAudioCodec(void* i2c_master_handle, int input_sample_rate, int
input_channels_ = input_reference_ ? 2 : 1; // 输入通道数
input_sample_rate_ = input_sample_rate;
output_sample_rate_ = output_sample_rate;
input_gain_ = 30;
CreateDuplexChannels(mclk, bclk, ws, dout, din);
@ -200,7 +201,7 @@ void BoxAudioCodec::EnableInput(bool enable) {
fs.channel_mask |= ESP_CODEC_DEV_MAKE_CHANNEL_MASK(1);
}
ESP_ERROR_CHECK(esp_codec_dev_open(input_dev_, &fs));
ESP_ERROR_CHECK(esp_codec_dev_set_in_channel_gain(input_dev_, ESP_CODEC_DEV_MAKE_CHANNEL_MASK(0), AUDIO_CODEC_DEFAULT_MIC_GAIN));
ESP_ERROR_CHECK(esp_codec_dev_set_in_channel_gain(input_dev_, ESP_CODEC_DEV_MAKE_CHANNEL_MASK(0), input_gain_));
} else {
ESP_ERROR_CHECK(esp_codec_dev_close(input_dev_));
}

3
main/audio/codecs/es8311_audio_codec.cc
View File

@ -14,6 +14,7 @@ Es8311AudioCodec::Es8311AudioCodec(void* i2c_master_handle, i2c_port_t i2c_port,
output_sample_rate_ = output_sample_rate;
pa_pin_ = pa_pin;
pa_inverted_ = pa_inverted;
input_gain_ = 30;
assert(input_sample_rate_ == output_sample_rate_);
CreateDuplexChannels(mclk, bclk, ws, dout, din);
@ -81,7 +82,7 @@ void Es8311AudioCodec::UpdateDeviceState() {
.mclk_multiple = 0,
};
ESP_ERROR_CHECK(esp_codec_dev_open(dev_, &fs));
ESP_ERROR_CHECK(esp_codec_dev_set_in_gain(dev_, AUDIO_CODEC_DEFAULT_MIC_GAIN));
ESP_ERROR_CHECK(esp_codec_dev_set_in_gain(dev_, input_gain_));
ESP_ERROR_CHECK(esp_codec_dev_set_out_vol(dev_, output_volume_));
} else if (!input_enabled_ && !output_enabled_ && dev_ != nullptr) {
esp_codec_dev_close(dev_);

4
main/audio/codecs/es8374_audio_codec.cc
View File

@ -12,6 +12,8 @@ Es8374AudioCodec::Es8374AudioCodec(void* i2c_master_handle, i2c_port_t i2c_port,
input_channels_ = 1; // 输入通道数
input_sample_rate_ = input_sample_rate;
output_sample_rate_ = output_sample_rate;
input_gain_ = 30;
pa_pin_ = pa_pin;
CreateDuplexChannels(mclk, bclk, ws, dout, din);
@ -146,7 +148,7 @@ void Es8374AudioCodec::EnableInput(bool enable) {
.mclk_multiple = 0,
};
ESP_ERROR_CHECK(esp_codec_dev_open(input_dev_, &fs));
ESP_ERROR_CHECK(esp_codec_dev_set_in_gain(input_dev_, AUDIO_CODEC_DEFAULT_MIC_GAIN));
ESP_ERROR_CHECK(esp_codec_dev_set_in_gain(input_dev_, input_gain_));
} else {
ESP_ERROR_CHECK(esp_codec_dev_close(input_dev_));
}

4
main/audio/codecs/es8388_audio_codec.cc
View File

@ -12,6 +12,8 @@ Es8388AudioCodec::Es8388AudioCodec(void* i2c_master_handle, i2c_port_t i2c_port,
input_channels_ = input_reference_ ? 2 : 1; // 输入通道数
input_sample_rate_ = input_sample_rate;
output_sample_rate_ = output_sample_rate;
input_gain_ = 24;
pa_pin_ = pa_pin;
CreateDuplexChannels(mclk, bclk, ws, dout, din);
@ -158,7 +160,7 @@ void Es8388AudioCodec::EnableInput(bool enable) {
uint8_t gain = (11 << 4) + 0;
ctrl_if_->write_reg(ctrl_if_, 0x09, 1, &gain, 1);
}else{
ESP_ERROR_CHECK(esp_codec_dev_set_in_gain(input_dev_, 24.0));
ESP_ERROR_CHECK(esp_codec_dev_set_in_gain(input_dev_, input_gain_));
}
} else {
ESP_ERROR_CHECK(esp_codec_dev_close(input_dev_));

3
main/audio/codecs/es8389_audio_codec.cc
View File

@ -12,6 +12,7 @@ Es8389AudioCodec::Es8389AudioCodec(void* i2c_master_handle, i2c_port_t i2c_port,
input_channels_ = 1; // 输入通道数
input_sample_rate_ = input_sample_rate;
output_sample_rate_ = output_sample_rate;
input_gain_ = 40;
pa_pin_ = pa_pin;
CreateDuplexChannels(mclk, bclk, ws, dout, din);
@ -153,7 +154,7 @@ void Es8389AudioCodec::EnableInput(bool enable) {
.mclk_multiple = 0,
};
ESP_ERROR_CHECK(esp_codec_dev_open(input_dev_, &fs));
ESP_ERROR_CHECK(esp_codec_dev_set_in_gain(input_dev_, 40.0));
ESP_ERROR_CHECK(esp_codec_dev_set_in_gain(input_dev_, input_gain_));
} else {
ESP_ERROR_CHECK(esp_codec_dev_close(input_dev_));
}

163
main/audio/codecs/no_audio_codec.cc
View File

@ -214,7 +214,53 @@ NoAudioCodecSimplex::NoAudioCodecSimplex(int input_sample_rate, int output_sampl
ESP_LOGI(TAG, "Simplex channels created");
}
NoAudioCodecSimplexPdm::NoAudioCodecSimplexPdm(int input_sample_rate, int output_sample_rate, gpio_num_t spk_bclk, gpio_num_t spk_ws, gpio_num_t spk_dout, i2s_std_slot_mask_t spk_slot_mask,gpio_num_t mic_sck, gpio_num_t mic_din) {
int NoAudioCodec::Write(const int16_t* data, int samples) {
std::lock_guard<std::mutex> lock(data_if_mutex_);
std::vector<int32_t> buffer(samples);
// output_volume_: 0-100
// volume_factor_: 0-65536
int32_t volume_factor = pow(double(output_volume_) / 100.0, 2) * 65536;
for (int i = 0; i < samples; i++) {
int64_t temp = int64_t(data[i]) * volume_factor; // 使用 int64_t 进行乘法运算
if (temp > INT32_MAX) {
buffer[i] = INT32_MAX;
} else if (temp < INT32_MIN) {
buffer[i] = INT32_MIN;
} else {
buffer[i] = static_cast<int32_t>(temp);
}
}
size_t bytes_written;
ESP_ERROR_CHECK(i2s_channel_write(tx_handle_, buffer.data(), samples * sizeof(int32_t), &bytes_written, portMAX_DELAY));
return bytes_written / sizeof(int32_t);
}
int NoAudioCodec::Read(int16_t* dest, int samples) {
size_t bytes_read;
std::vector<int32_t> bit32_buffer(samples);
if (i2s_channel_read(rx_handle_, bit32_buffer.data(), samples * sizeof(int32_t), &bytes_read, portMAX_DELAY) != ESP_OK) {
ESP_LOGE(TAG, "Read Failed!");
return 0;
}
samples = bytes_read / sizeof(int32_t);
for (int i = 0; i < samples; i++) {
int32_t value = bit32_buffer[i] >> 12;
dest[i] = (value > INT16_MAX) ? INT16_MAX : (value < -INT16_MAX) ? -INT16_MAX : (int16_t)value;
}
return samples;
}
// Delegating constructor: calls the main constructor with default slot mask
NoAudioCodecSimplexPdm::NoAudioCodecSimplexPdm(int input_sample_rate, int output_sample_rate, gpio_num_t spk_bclk, gpio_num_t spk_ws, gpio_num_t spk_dout, gpio_num_t mic_sck, gpio_num_t mic_din)
: NoAudioCodecSimplexPdm(input_sample_rate, output_sample_rate, spk_bclk, spk_ws, spk_dout, I2S_STD_SLOT_LEFT, mic_sck, mic_din) {
// All initialization is handled by the delegated constructor
}
NoAudioCodecSimplexPdm::NoAudioCodecSimplexPdm(int input_sample_rate, int output_sample_rate, gpio_num_t spk_bclk, gpio_num_t spk_ws, gpio_num_t spk_dout, i2s_std_slot_mask_t spk_slot_mask, gpio_num_t mic_sck, gpio_num_t mic_din) {
duplex_ = false;
input_sample_rate_ = input_sample_rate;
output_sample_rate_ = output_sample_rate;
@ -292,110 +338,6 @@ NoAudioCodecSimplexPdm::NoAudioCodecSimplexPdm(int input_sample_rate, int output
ESP_LOGI(TAG, "Simplex channels created");
}
NoAudioCodecSimplexPdm::NoAudioCodecSimplexPdm(int input_sample_rate, int output_sample_rate, gpio_num_t spk_bclk, gpio_num_t spk_ws, gpio_num_t spk_dout, gpio_num_t mic_sck, gpio_num_t mic_din) {
duplex_ = false;
input_sample_rate_ = input_sample_rate;
output_sample_rate_ = output_sample_rate;
// Create a new channel for speaker
i2s_chan_config_t tx_chan_cfg = I2S_CHANNEL_DEFAULT_CONFIG((i2s_port_t)1, I2S_ROLE_MASTER);
tx_chan_cfg.dma_desc_num = AUDIO_CODEC_DMA_DESC_NUM;
tx_chan_cfg.dma_frame_num = AUDIO_CODEC_DMA_FRAME_NUM;
tx_chan_cfg.auto_clear_after_cb = true;
tx_chan_cfg.auto_clear_before_cb = false;
tx_chan_cfg.intr_priority = 0;
ESP_ERROR_CHECK(i2s_new_channel(&tx_chan_cfg, &tx_handle_, NULL));
i2s_std_config_t tx_std_cfg = {
.clk_cfg = {
.sample_rate_hz = (uint32_t)output_sample_rate_,
.clk_src = I2S_CLK_SRC_DEFAULT,
.mclk_multiple = I2S_MCLK_MULTIPLE_256,
#ifdef I2S_HW_VERSION_2
.ext_clk_freq_hz = 0,
#endif
},
.slot_cfg = I2S_STD_MSB_SLOT_DEFAULT_CONFIG(I2S_DATA_BIT_WIDTH_32BIT, I2S_SLOT_MODE_MONO),
.gpio_cfg = {
.mclk = I2S_GPIO_UNUSED,
.bclk = spk_bclk,
.ws = spk_ws,
.dout = spk_dout,
.din = I2S_GPIO_UNUSED,
.invert_flags = {
.mclk_inv = false,
.bclk_inv = false,
.ws_inv = false,
},
},
};
ESP_ERROR_CHECK(i2s_channel_init_std_mode(tx_handle_, &tx_std_cfg));
#if SOC_I2S_SUPPORTS_PDM_RX
// Create a new channel for MIC in PDM mode
i2s_chan_config_t rx_chan_cfg = I2S_CHANNEL_DEFAULT_CONFIG((i2s_port_t)0, I2S_ROLE_MASTER);
ESP_ERROR_CHECK(i2s_new_channel(&rx_chan_cfg, NULL, &rx_handle_));
i2s_pdm_rx_config_t pdm_rx_cfg = {
.clk_cfg = I2S_PDM_RX_CLK_DEFAULT_CONFIG((uint32_t)input_sample_rate_),
/* The data bit-width of PDM mode is fixed to 16 */
.slot_cfg = I2S_PDM_RX_SLOT_DEFAULT_CONFIG(I2S_DATA_BIT_WIDTH_16BIT, I2S_SLOT_MODE_MONO),
.gpio_cfg = {
.clk = mic_sck,
.din = mic_din,
.invert_flags = {
.clk_inv = false,
},
},
};
ESP_ERROR_CHECK(i2s_channel_init_pdm_rx_mode(rx_handle_, &pdm_rx_cfg));
#else
ESP_LOGE(TAG, "PDM is not supported");
#endif
ESP_LOGI(TAG, "Simplex channels created");
}
int NoAudioCodec::Write(const int16_t* data, int samples) {
std::lock_guard<std::mutex> lock(data_if_mutex_);
std::vector<int32_t> buffer(samples);
// output_volume_: 0-100
// volume_factor_: 0-65536
int32_t volume_factor = pow(double(output_volume_) / 100.0, 2) * 65536;
for (int i = 0; i < samples; i++) {
int64_t temp = int64_t(data[i]) * volume_factor; // 使用 int64_t 进行乘法运算
if (temp > INT32_MAX) {
buffer[i] = INT32_MAX;
} else if (temp < INT32_MIN) {
buffer[i] = INT32_MIN;
} else {
buffer[i] = static_cast<int32_t>(temp);
}
}
size_t bytes_written;
ESP_ERROR_CHECK(i2s_channel_write(tx_handle_, buffer.data(), samples * sizeof(int32_t), &bytes_written, portMAX_DELAY));
return bytes_written / sizeof(int32_t);
}
int NoAudioCodec::Read(int16_t* dest, int samples) {
size_t bytes_read;
std::vector<int32_t> bit32_buffer(samples);
if (i2s_channel_read(rx_handle_, bit32_buffer.data(), samples * sizeof(int32_t), &bytes_read, portMAX_DELAY) != ESP_OK) {
ESP_LOGE(TAG, "Read Failed!");
return 0;
}
samples = bytes_read / sizeof(int32_t);
for (int i = 0; i < samples; i++) {
int32_t value = bit32_buffer[i] >> 12;
dest[i] = (value > INT16_MAX) ? INT16_MAX : (value < -INT16_MAX) ? -INT16_MAX : (int16_t)value;
}
return samples;
}
int NoAudioCodecSimplexPdm::Read(int16_t* dest, int samples) {
size_t bytes_read;
@ -405,6 +347,13 @@ int NoAudioCodecSimplexPdm::Read(int16_t* dest, int samples) {
return 0;
}
// 计算实际读取的样本数
return bytes_read / sizeof(int16_t);
samples = bytes_read / sizeof(int16_t);
if (input_gain_ > 0) {
int gain_factor = (int)input_gain_;
for (int i = 0; i < samples; i++) {
int32_t amplified = dest[i] * gain_factor;
dest[i] = (amplified > INT16_MAX) ? INT16_MAX : (amplified < -INT16_MAX) ? -INT16_MAX : (int16_t)amplified;
}
}
return samples;
}

3
main/boards/m5stack-core-s3/cores3_audio_codec.cc
View File

@ -14,6 +14,7 @@ CoreS3AudioCodec::CoreS3AudioCodec(void* i2c_master_handle, int input_sample_rat
input_channels_ = input_reference_ ? 2 : 1; // 输入通道数
input_sample_rate_ = input_sample_rate;
output_sample_rate_ = output_sample_rate;
input_gain_ = 30;
CreateDuplexChannels(mclk, bclk, ws, dout, din);
@ -200,7 +201,7 @@ void CoreS3AudioCodec::EnableInput(bool enable) {
fs.channel_mask |= ESP_CODEC_DEV_MAKE_CHANNEL_MASK(1);
}
ESP_ERROR_CHECK(esp_codec_dev_open(input_dev_, &fs));
ESP_ERROR_CHECK(esp_codec_dev_set_in_channel_gain(input_dev_, ESP_CODEC_DEV_MAKE_CHANNEL_MASK(0), AUDIO_CODEC_DEFAULT_MIC_GAIN));
ESP_ERROR_CHECK(esp_codec_dev_set_in_channel_gain(input_dev_, ESP_CODEC_DEV_MAKE_CHANNEL_MASK(0), input_gain_));
} else {
ESP_ERROR_CHECK(esp_codec_dev_close(input_dev_));
}

3
main/boards/m5stack-tab5/tab5_audio_codec.cc
View File

@ -14,6 +14,7 @@ Tab5AudioCodec::Tab5AudioCodec(void* i2c_master_handle, int input_sample_rate, i
input_channels_ = input_reference_ ? 2 : 1; // 输入通道数
input_sample_rate_ = input_sample_rate;
output_sample_rate_ = output_sample_rate;
input_gain_ = 30;
CreateDuplexChannels(mclk, bclk, ws, dout, din);
@ -199,7 +200,7 @@ void Tab5AudioCodec::EnableInput(bool enable) {
fs.channel_mask |= ESP_CODEC_DEV_MAKE_CHANNEL_MASK(1);
}
ESP_ERROR_CHECK(esp_codec_dev_open(input_dev_, &fs));
ESP_ERROR_CHECK(esp_codec_dev_set_in_channel_gain(input_dev_, ESP_CODEC_DEV_MAKE_CHANNEL_MASK(0), AUDIO_CODEC_DEFAULT_MIC_GAIN));
ESP_ERROR_CHECK(esp_codec_dev_set_in_channel_gain(input_dev_, ESP_CODEC_DEV_MAKE_CHANNEL_MASK(0), input_gain_));
} else {
ESP_ERROR_CHECK(esp_codec_dev_close(input_dev_));
}