Then buy an ESP32-S3-Box-3 when mouser releases them https://www.mouser.com/ProductDetail/Espressif-Systems/ESP32-S3-BOX-3?qs=sGAEpiMZZMuqBwn8WqcFUipNgoezRlc4u4gjCc9cblNcfZiByE2ecw%3D%3D
1 Like
yes but I want also to use them to stream music (so the speaker have to be “solid”)
There were some discussions in this thread as to whether or not a bluetooth speaker with built-in mic could be used directly as a satellite (without esp etc.). Did anything ever come of that or is there a thread for this approach?
I love the creativity of this and some of the other projects here 
but ultimately I would also be interested in just using out of the box bluetooth devices by simply pairing them to the HA server or an esp32 acting solely as a bluetooth to wifi relay.
And that will work as a satellite that reacts to wake words?
I dont think anyone has gotten that to work yet.
Been googling this but can’t see to find how to do it. Any pointer?
It just acts like a plain microphone / speaker for the system whatever you have on that system should be able to use it as such.
Debian Bullseye (Skip this and upgrade to bookworm directions below it works better)
sudo apt update
sudo apt full-upgrade
sudo apt -y dist-upgrade
sudo reboot
Check to see if SAP failed, and if so remove the plugin:
sudo systemctl status bluetooth
● bluetooth.service - Bluetooth service
Loaded: loaded (/lib/systemd/system/bluetooth.service; enabled; vendor preset: enabled)
Active: active (running) since Sat 2024-03-09 19:15:27 EST; 5min ago
Docs: man:bluetoothd(8)
Main PID: 4016 (bluetoothd)
Status: "Running"
Tasks: 1 (limit: 779)
CPU: 166ms
CGroup: /system.slice/bluetooth.service
└─4016 /usr/libexec/bluetooth/bluetoothd
Mar 09 19:15:27 VA-001 bluetoothd[4016]: Bluetooth daemon 5.55
Mar 09 19:15:27 VA-001 systemd[1]: Started Bluetooth service.
Mar 09 19:15:27 VA-001 bluetoothd[4016]: Starting SDP server
Mar 09 19:15:27 VA-001 bluetoothd[4016]: Bluetooth management interface 1.22 initialized
Mar 09 19:15:27 VA-001 bluetoothd[4016]: profiles/sap/server.c:sap_server_register() Sap driver initialization failed.
Mar 09 19:15:27 VA-001 bluetoothd[4016]: sap-server: Operation not permitted (1)
Mar 09 19:15:27 VA-001 bluetoothd[4016]: Failed to set mode: Failed (0x03)
Mar 09 19:15:27 VA-001 bluetoothd[4016]: Set device flags return status: Invalid Parameters
Mar 09 19:15:27 VA-001 bluetoothd[4016]: Endpoint registered: sender=:1.30 path=/MediaEndpoint/A2DPSink/sbc
Mar 09 19:15:27 VA-001 bluetoothd[4016]: Endpoint registered: sender=:1.30 path=/MediaEndpoint/A2DPSource/sbc
sudo nano /lib/systemd/system/bluetooth.service
Edit the ExecStart=/usr/lib/bluetooth/bluetoothd line to add --noplugin=sap
Should look like this
ExecStart=/usr/libexec/bluetooth/bluetoothd --noplugin=sap
Ctrl + O, Ctrl + X to write then exit
sudo systemctl daemon-reload
sudo systemctl restart bluetooth
sudo apt -y install --no-install-recommends pulseaudio-module-bluetooth
sudo nano /etc/systemd/system/pulseaudio.service
[Unit]
Description=Pulseaudio sound server
After=avahi-daemon.service network.target
[Service]
ExecStart=/usr/bin/pulseaudio --system --disallow-exit --disable-shm --daemonize=no
ExecReload=/bin/kill -HUP $MAINPID
[Install]
WantedBy=multi-user.target
Save/Exit
Add this to the end of the system.pa file:
sudo nano /etc/pulse/system.pa
.ifexists module-bluetooth-policy.so
load-module module-bluetooth-policy
.endif
.ifexists module-bluetooth-discover.so
load-module module-bluetooth-discover
.endif
Save/Exit
sudo adduser pulse bluetooth
Adding user `pulse' to group `bluetooth' ...
Adding user pulse to group bluetooth
Done.
sudo adduser pi pulse-access
Adding user 'pi' to group `pulse-access' ...
Adding user pi to group pulse-access
Done.
Add this to the end of the default.pa file:
sudo nano /etc/pulse/default.pa
# automatically switch to newly-connected devices
load-module module-switch-on-connect
sudo systemctl restart dbus
sudo systemctl enable pulseaudio
sudo systemctl start pulseaudio
sudo usermod -G bluetooth -a pi
su - $USER
id
rfkill list
0: phy0: Wireless LAN
Soft blocked: no
Hard blocked: no
1: hci0: Bluetooth
Soft blocked: yes
Hard blocked: no
rfkill unblock all
rfkill list
0: phy0: Wireless LAN
Soft blocked: no
Hard blocked: no
1: hci0: Bluetooth
Soft blocked: no
Hard blocked: no
bluetoothctl
Agent registered
power on
Changing power on succeeded
agent on
Agent is already registered
default-agent
Default agent request successful
scan on
Discovery started
[CHG] Controller DE:AD:BE:EF:CA:FE Discovering: yes
[NEW] Device DE:AD:BE:EF:FA:CE AudioHeadSet
pair de:ad:be:ef:fa:ce
Attempting to pair with DE:AD:BE:EF:FA:CE
[CHG] Device DE:AD:BE:EF:FA:CE Connected: yes
[CHG] Device DE:AD:BE:EF:FA:CE UUIDs: #########-####-####-####-###############
[CHG] Device DE:AD:BE:EF:FA:CE UUIDs: #########-####-####-####-###############
[CHG] Device DE:AD:BE:EF:FA:CE ServicesResolved: yes
[CHG] Device DE:AD:BE:EF:FA:CE Paired: yes
Pairing successful
[CHG] Device DE:AD:BE:EF:FA:CE ServicesResolved: no
[CHG] Device DE:AD:BE:EF:FA:CE Connected: no
trust de:ad:be:ef:fa:ce
[CHG] Device DE:AD:BE:EF:FA:CE Trusted: yes
Changing DE:AD:BE:EF:FA:CE trust succeeded
connect de:ad:be:ef:fa:ce
Attempting to connect to DE:AD:BE:EF:FA:CE
[CHG] Device DE:AD:BE:EF:FA:CE Connected: yes
Connection successful
[CHG] Device DE:AD:BE:EF:FA:CE ServicesResolved: yes
scan off
quit
sudo reboot
LANG=C pactl list | grep -A2 'Source #' | grep 'Name: ' | cut -d" " -f2
alsa_output.platform-bcm2835_audio.analog-stereo.monitor
bluez_sink.DE_AD_BE_EF_FA_CE.a2dp_sink.monitor
aplay /usr/share/sounds/alsa/Front_Center.wav
Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit Little Endian, Rate 48000 Hz, Mono
arecord -d 5 test.wav
aplay test.wav
General Satellite Setup
sudo apt-get install python3 python3-pip python3-venv \
alsa-utils git
sudo apt-get install --no-install-recommends \
ffmpeg
cd
mkdir code
cd code
git clone https://github.com/synesthesiam/homeassistant-satellite.git
cd homeassistant-satellite
script/setup
sudo nano /etc/systemd/system/homeassistant-satellite.service
[Unit]
Description=Home Assistant Satellite
Wants=network-online.target
After=network-online.target
[Service]
Type=simple
ExecStart=/home/pi/code/homeassistant-satellite/script/run --host 10.0.1.42 --token <HALongLiveTokeHere> --awake-sound sounds/awake.wav --done-sound sounds/done.wav --mic-device pulse --snd-device pulse
WorkingDirectory=/home/pi/code/homeassistant-satellite
Restart=always
RestartSec=1
[Install]
WantedBy=default.target
Debian Bookworm and Pipewire the bluetooth mic works using Profile: Headset Head Unit (HSP/HFP, codec CVSD)
sudo apt install pipewire \
libspa-0.2-bluetooth \
pipewire-audio-client-libraries \
pipewire-media-session- \
wireplumber \
bluez \
blueman \
pavucontrol
sudo usermod -G bluetooth -a pi
su - $USER
id
rfkill list
0: phy0: Wireless LAN
Soft blocked: no
Hard blocked: no
1: hci0: Bluetooth
Soft blocked: yes
Hard blocked: no
rfkill unblock all
rfkill list
0: phy0: Wireless LAN
Soft blocked: no
Hard blocked: no
1: hci0: Bluetooth
Soft blocked: no
Hard blocked: no
bluetoothctl
Agent registered
power on
Changing power on succeeded
agent on
Agent is already registered
default-agent
Default agent request successful
scan on
Discovery started
[CHG] Controller DE:AD:BE:EF:CA:FE Discovering: yes
[NEW] Device DE:AD:BE:EF:FA:CE AudioHeadSet
pair de:ad:be:ef:fa:ce
Attempting to pair with DE:AD:BE:EF:FA:CE
[CHG] Device DE:AD:BE:EF:FA:CE Connected: yes
[CHG] Device DE:AD:BE:EF:FA:CE UUIDs: #########-####-####-####-###############
[CHG] Device DE:AD:BE:EF:FA:CE UUIDs: #########-####-####-####-###############
[CHG] Device DE:AD:BE:EF:FA:CE ServicesResolved: yes
[CHG] Device DE:AD:BE:EF:FA:CE Paired: yes
Pairing successful
[CHG] Device DE:AD:BE:EF:FA:CE ServicesResolved: no
[CHG] Device DE:AD:BE:EF:FA:CE Connected: no
trust de:ad:be:ef:fa:ce
[CHG] Device DE:AD:BE:EF:FA:CE Trusted: yes
Changing DE:AD:BE:EF:FA:CE trust succeeded
connect de:ad:be:ef:fa:ce
Attempting to connect to DE:AD:BE:EF:FA:CE
[CHG] Device DE:AD:BE:EF:FA:CE Connected: yes
Connection successful
[CHG] Device DE:AD:BE:EF:FA:CE ServicesResolved: yes
scan off
quit
If HSP is not showing up:
sudo cp -r /var/lib/bluetooth /var/lib/bluetooth_BACKUP
sudo systemctl stop bluetooth
sudo rm -rf /var/lib/bluetooth/*
sudo systemctl start bluetooth
sudo reboot
sudo apt install gstreamer1.0
gst-launch-1.0 audiotestsrc ! audioresample ! autoaudiosink
Setting pipeline to PAUSED ...
Pipeline is PREROLLING ...
Redistribute latency...
Pipeline is PREROLLED ...
Setting pipeline to PLAYING ...
Redistribute latency...
New clock: GstPulseSinkClock
0:00:00.1 / 99:99:99.
You should hear a beep.
pw-cli list-objects | grep node.name
node.name = "Dummy-Driver"
node.name = "Freewheel-Driver"
node.name = "Midi-Bridge"
node.name = "v4l2_input.platform-bcm2835-isp.2"
node.name = "v4l2_input.platform-bcm2835-isp.3"
node.name = "v4l2_input.platform-bcm2835-isp.6"
node.name = "v4l2_input.platform-bcm2835-isp.7"
node.name = "alsa_output.platform-bcm2835_audio.stereo-fallback"
node.name = "bluez_input.DE_AD_BE_EF_FA_CE.0"
node.name = "bluez_output.DE_AD_BE_EF_FA_CE.1"
You can switch between the AD2P and HSP profile like this:
pactl set-card-profile bluez_card.DE_AD_BE_EF_FA_CE headset-head-unit-cvsd
pactl set-card-profile bluez_card.DE_AD_BE_EF_FA_CE a2dp-sink-sbc
pactl list
git clone https://github.com/rhasspy/wyoming-openwakeword.git
cd wyoming-openwakeword
script/setup
script/run --uri 'tcp://0.0.0.0:10400' --debug --preload-model 'hey_jarvis'
git clone https://github.com/rhasspy/wyoming-satellite.git
cd wyoming-satellite.git
scripts/setup
script/run --name 'bluetooth satellite' --uri 'tcp://0.0.0.0:10700' --mic-command 'parecord --device bluez_input.DE_AD_BE_EF_FA_CE.0 --rate=16000 --channels=1 --format=s16le --raw' --snd-command 'paplay --device bluez_output.DE_AD_BE_EF_FA_CE.1 --rate=22000 --channels=1 --format=s16le --raw' --debug --wake-uri 'tcp://127.0.0.1:10400' --wake-word-name 'hey_jarvis' --done-wav sounds/done.wav --awake-wav sounds/awake.wav
INFO:root:Ready
DEBUG:root:Detected IP: 10.0.42.100
DEBUG:root:Zeroconf discovery enabled (name=############, host=None)
DEBUG:root:Connecting to mic service: ['parecord', '--device', 'bluez_input.DE_AD_BE_EF_FA_CE.0', '--rate=16000', '--channels=1', '--format=s16le', '--raw']
DEBUG:root:Connecting to snd service: ['paplay', '--device', 'bluez_output.DE_AD_BE_EF_FA_CE.1', '--rate=16000', '--channels=1', '--format=s16le', '--raw']
DEBUG:root:Connecting to wake service: tcp://127.0.0.1:10400
INFO:root:Connected to services
DEBUG:root:Connected to mic service
DEBUG:root:Connected to wake service
DEBUG:root:Server set: 3667213960213
INFO:root:Connected to server
INFO:root:Waiting for wake word
DEBUG:root:Ping enabled
DEBUG:root:Detection(name='hey_jarvis_v0.1', timestamp=3674454417437)
DEBUG:root:Streaming audio
DEBUG:root:Event(type='run-pipeline', data={'start_stage': 'asr', 'end_stage': 'tts', 'restart_on_end': False, 'snd_format': {'rate': 22050, 'width': 2, 'channels': 1}}, payload=None)
DEBUG:root:Muting microphone for 0.8995918367346939 second(s)
DEBUG:root:Connected to snd service
DEBUG:root:Unmuted microphone
DEBUG:root:Connected to snd service
DEBUG:root:Unmuted microphone
DEBUG:root:Event(type='transcript', data={'text': ' Turn off the family room light.'}, payload=None)
INFO:root:Waiting for wake word
DEBUG:root:Connected to snd service
DEBUG:root:Event(type='synthesize', data={'text': 'Turned off the lights', 'voice': {'name': 'en_US-lessac-medium'}}, payload=None)
DEBUG:root:Connected to snd service
DEBUG:root:Detection(name='hey_jarvis_v0.1', timestamp=3784489216895)
DEBUG:root:Streaming audio
1 Like
Here is my esphome config for the esspressif esp32s3 korvo-1 development board. I ended up getting it since an S3 Box is impossible to get. It works, all the LED’s work as they should but it could probably use some tweaking to take full advantage of the hardware. Now I just need to design a 3D case. Best part is it has speaker outputs for low power or powered speakers and a 3.5mm output for pretty much anything you want to hook it into for sound output. No onboard speaker though.
substitutions:
friendly_name: korvo
esphome:
name: korvo
friendly_name: ${friendly_name}
name_add_mac_suffix: true
platformio_options:
board_build.flash_mode: dio
upload_speed: 460800
project:
name: esphome.voice-assistant
version: "1.0"
min_version: 2023.11.1
on_boot:
- priority: 600
then:
- light.turn_on:
id: led_ring
brightness: 70%
effect: connecting
esp32:
board: esp32-s3-devkitc-1
framework:
type: esp-idf
sdkconfig_options:
CONFIG_ESP32S3_DEFAULT_CPU_FREQ_240: "y"
CONFIG_ESP32S3_DATA_CACHE_64KB: "y"
CONFIG_ESP32S3_DATA_CACHE_LINE_64B: "y"
CONFIG_AUDIO_BOARD_CUSTOM: "y"
CONFIG_ESP32_S3_KORVO1_BOARD: "y"
components:
- name: esp32_s3_korvo1_board
source: github://abmantis/esphome_custom_audio_boards@main
refresh: 0s
psram:
mode: octal
speed: 80MHz
external_components:
- source: github://pr#5230
components: esp_adf
refresh: 0s
ota:
logger:
api:
on_client_connected:
then:
- if:
condition:
switch.is_on: use_wake_word
then:
- delay: 1s
- voice_assistant.start_continuous:
- delay: 1s
- voice_assistant.stop:
- delay: 2s
- voice_assistant.start_continuous:
- script.execute: reset_led
on_client_disconnected:
then:
- light.turn_on:
id: led_ring
blue: 0%
red: 100%
green: 100%
brightness: 50%
effect: connecting
dashboard_import:
package_import_url: github://esphome/firmware/voice-assistant/esp32-s3-korvo1.yaml@main
wifi:
use_address: 192.168.0.57
ap:
on_connect:
then:
- delay: 5s # Gives time for improv results to be transmitted
- ble.disable:
on_disconnect:
then:
- ble.enable:
improv_serial:
esp32_improv:
authorizer: none
button:
- platform: factory_reset
id: factory_reset_btn
name: Factory reset
esp_adf:
board: esp32s3korvo1
microphone:
- platform: esp_adf
id: korvo_mic
speaker:
- platform: esp_adf
id: korvo_speaker
voice_assistant:
id: voice_asst
microphone: korvo_mic
speaker: korvo_speaker
noise_suppression_level: 4
auto_gain: 10dBFS
volume_multiplier: 1
use_wake_word: false
on_listening:
- light.turn_on:
id: led_ring
blue: 100%
red: 0%
green: 0%
brightness: 100%
effect: wakeword
on_tts_start:
- light.turn_on:
id: led_ring
blue: 0%
red: 0%
green: 100%
brightness: 50%
effect: pulse
on_end:
- delay: 100ms
- wait_until:
not:
speaker.is_playing:
- script.execute: reset_led
on_error:
- light.turn_on:
id: led_ring
blue: 0%
red: 100%
green: 0%
brightness: 100%
effect: none
- delay: 1s
- script.execute: reset_led
- script.wait: reset_led
- lambda: |-
if (code == "wake-provider-missing" || code == "wake-engine-missing") {
id(use_wake_word).turn_off();
}
script:
- id: reset_led
then:
- if:
condition:
switch.is_on: use_wake_word
then:
- light.turn_on:
id: led_ring
blue: 100%
red: 0%
green: 0%
brightness: 30%
effect: none
else:
- light.turn_off: led_ring
switch:
- platform: gpio
id: pa_ctrl
pin: GPIO38
name: "${friendly_name} Speaker Mute"
restore_mode: ALWAYS_ON
- platform: template
name: Use wake word
id: use_wake_word
optimistic: true
restore_mode: RESTORE_DEFAULT_ON
entity_category: config
on_turn_on:
- lambda: id(voice_asst).set_use_wake_word(true);
- if:
condition:
not:
- voice_assistant.is_running
then:
- voice_assistant.start_continuous
- script.execute: reset_led
on_turn_off:
- voice_assistant.stop
- script.execute: reset_led
light:
- platform: esp32_rmt_led_strip
id: led_ring
name: "${friendly_name} Light"
pin: GPIO19
num_leds: 12
rmt_channel: 0
rgb_order: GRB
chipset: ws2812
default_transition_length: 0s
effects:
- pulse:
name: "Pulse"
transition_length: 0.5s
update_interval: 0.5s
- addressable_twinkle:
name: "Working"
twinkle_probability: 5%
progress_interval: 4ms
- addressable_color_wipe:
name: "Wakeword"
colors:
- red: 0%
green: 50%
blue: 0%
num_leds: 12
add_led_interval: 20ms
reverse: false
- addressable_color_wipe:
name: "Connecting"
colors:
- red: 60%
green: 60%
blue: 60%
num_leds: 12
- red: 60%
green: 60%
blue: 0%
num_leds: 12
add_led_interval: 100ms
reverse: true
binary_sensor:
- platform: template
name: "${friendly_name} Volume Up"
id: btn_volume_up
- platform: template
name: "${friendly_name} Volume Down"
id: btn_volume_down
- platform: template
name: "${friendly_name} Set"
id: btn_set
- platform: template
name: "${friendly_name} Play"
id: btn_play
- platform: template
name: "${friendly_name} Mode"
id: btn_mode
on_multi_click:
- timing:
- ON for at least 10s
then:
- button.press: factory_reset_btn
- platform: template
name: "${friendly_name} Record"
id: btn_record
on_press:
- voice_assistant.start:
- light.turn_on:
id: led_ring
brightness: 100%
effect: "Wakeword"
on_release:
- voice_assistant.stop:
- light.turn_off:
id: led_ring
sensor:
- id: button_adc
platform: adc
internal: true
pin: 8
attenuation: 11db
update_interval: 15ms
filters:
- median:
window_size: 5
send_every: 5
send_first_at: 1
- delta: 0.1
on_value_range:
- below: 0.55
then:
- binary_sensor.template.publish:
id: btn_volume_up
state: ON
- above: 0.65
below: 0.92
then:
- binary_sensor.template.publish:
id: btn_volume_down
state: ON
- above: 1.02
below: 1.33
then:
- binary_sensor.template.publish:
id: btn_set
state: ON
- above: 1.43
below: 1.77
then:
- binary_sensor.template.publish:
id: btn_play
state: ON
- above: 1.87
below: 2.15
then:
- binary_sensor.template.publish:
id: btn_mode
state: ON
- above: 2.25
below: 2.56
then:
- binary_sensor.template.publish:
id: btn_record
state: ON
- above: 2.8
then:
- binary_sensor.template.publish:
id: btn_volume_up
state: OFF
- binary_sensor.template.publish:
id: btn_volume_down
state: OFF
- binary_sensor.template.publish:
id: btn_set
state: OFF
- binary_sensor.template.publish:
id: btn_play
state: OFF
- binary_sensor.template.publish:
id: btn_mode
state: OFF
- binary_sensor.template.publish:
id: btn_record
state: OFF
1 Like
For anyone looking for an elegant, 3d printed (that is covered with a textile speaker grille) enclosure for the Korvo and Korvo1, have a look at my WIP over here Far field satellite with an Elegant 3d printed enclosures
I’ve been following this thread, but it would be great if someone could wrap it up, with some simple to follow instructions on the board in regards to how it all finalized with what works best. both for the S3 dev board and the regular esp board.
1 Like
Hello, with my ESP32 N16R8 I was able to control the RGB LED, use the wake word, the actions are well done but despite all my attempts I did not manage to output sound via the Max98357 either in speaker mode or media_player. So also a taker of a connection / code that works with the ESP32 S3 
Here is a link to the code I use on N16R8. It works, but has a few false positives when the radio is on in the background. So need to do some fiddling about with settings.
Thanks to BigBobbas
As an update to my crackling speaker issues, I replaced the max98357 and it was fine. So one non working amp.
1 Like
hi
i seen your post, i tried the same setup with my esp32 wroom32 but have issues with the media_player, when i play a file the esp32 crashes.
is the ESP32-s3-N16R8 Dev Board better than the “ESP32 wroom 32” ?
how do you make it which the command to stop the music?
i read something about custom commands with the assist, or is it possible to mute the sound when the wake word is speaken?`
Everyone with noise problems… it is the 3.3V or 5V supply voltage… I run an OrangePI 2W with USB Audio and it was noisy. The solution is to Boost 5V to 8V and use an analog Voltage regulator such as 7805. I cut the red and black cables in an USB extention cable being careful of the data wires, and connected the MT3608 Boost converter and the regulator 7805 supplying smooth voltage to the USB audio module. Be aware if you need 3.3V you need a 3.3V regulator and no Boost converter if you have 5V somewhere. For me the noise disapeared completely /Mikael
3 Likes
Well, I was wrong, i thought the repository was handling microwakeword and it wasn’t, here is the updated working esphome code for microwakeword. Also, if you flashed it with the board as esp32-se-devkitc-1, it only formats as either 4MB or 8MB, I was pulling my hair out until I realized the PSRAM was more then double the ROM showing in esphome. It said the BIN file was 19xxx bytes and the flash was 18xxx bytes so I automatically assumed 18MB when it’s nowhere close to that… Just had to specify the flash size, one line…
Remeber to change API key, and wifi as I have it statically IP’d. Also, it points to a non existent file for the firmware, because of this, it shows up as red in esphome but everything works and you will always be prompted to install a “new” device. Just leave it alone, don’t adopt it. From what I understand, it will need to be added on the esphome side and there has been a feature request to add it. If that will happen or when, if ever, I have no idea. If you comment it out, it fails to flash… Everything works without issues although I’m pretty postie the buttons don’t do anything. There are some warning during install and they are way above my skill set as I am not a developer but can cut and paste other peoples code and mess around with it and usually can get things to work. Most credit goes out to whoever wrote the repository I am pointing to because I never would have been able to do it without that
substitutions:
name: "korvo"
friendly_name: korvo
esphome:
name: ${name}
friendly_name: ${friendly_name}
name_add_mac_suffix: true
platformio_options:
board_build.flash_mode: dio
upload_speed: 460800
project:
name: esphome.voice-assistant
version: "1.0"
min_version: 2023.11.5
on_boot:
- priority: 600
then:
- light.turn_on:
id: led_ring
brightness: 70%
effect: connecting
esp32:
board: esp32-s3-devkitc-1
flash_size: 16MB
framework:
type: esp-idf
sdkconfig_options:
CONFIG_ESP32S3_DEFAULT_CPU_FREQ_240: "y"
CONFIG_ESP32S3_DATA_CACHE_64KB: "y"
CONFIG_ESP32S3_DATA_CACHE_LINE_64B: "y"
CONFIG_AUDIO_BOARD_CUSTOM: "y"
CONFIG_ESP32_S3_KORVO1_BOARD: "y"
components:
- name: esp32_s3_korvo1_board
source: github://abmantis/esphome_custom_audio_boards@main
refresh: 0s
psram:
mode: octal
speed: 80MHz
external_components:
- source: github://pr#5230
components: esp_adf
refresh: 0s
ota:
logger:
api:
encryption:
key: you_api_key
on_client_connected:
then:
- if:
condition:
switch.is_on: use_wake_word
then:
- delay: 1s
- voice_assistant.start_continuous:
- delay: 1s
- voice_assistant.stop:
- delay: 1500ms
- voice_assistant.start_continuous:
- script.execute: reset_led
on_client_disconnected:
then:
- light.turn_on:
id: led_ring
blue: 0%
red: 100%
green: 100%
brightness: 50%
effect: connecting
dashboard_import:
package_import_url: github://esphome/firmware/voice-assistant/esp32-s3-korvo1.yaml@main
wifi:
ssid: !secret wifi_ssid
password: !secret wifi_password
use_address: 192.168.0.xx
ap:
on_connect:
then:
- delay: 5s # Gives time for improv results to be transmitted
- ble.disable:
on_disconnect:
then:
- ble.enable:
improv_serial:
esp32_improv:
authorizer: none
button:
- platform: factory_reset
id: factory_reset_btn
name: Factory reset
esp_adf:
board: esp32s3korvo1
microphone:
- platform: esp_adf
id: korvo_mic
speaker:
- platform: esp_adf
id: korvo_speaker
micro_wake_word:
on_wake_word_detected:
# then:
- voice_assistant.start:
wake_word: !lambda return wake_word;
- light.turn_on:
id: led_ring
red: 30%
green: 30%
blue: 70%
brightness: 60%
effect: fast pulse
model: okay_nabu.
voice_assistant:
id: voice_asst
microphone: korvo_mic
speaker: korvo_speaker
noise_suppression_level: 2
auto_gain: 31dBFS
volume_multiplier: 2
use_wake_word: true
on_listening:
- light.turn_on:
id: led_ring
blue: 100%
red: 0%
green: 0%
brightness: 100%
effect: wakeword
on_tts_start:
- light.turn_on:
id: led_ring
blue: 0%
red: 0%
green: 100%
brightness: 50%
effect: pulse
on_end:
- delay: 500ms
- wait_until:
not:
speaker.is_playing:
- script.execute: reset_led
on_error:
- light.turn_on:
id: led_ring
blue: 0%
red: 100%
green: 0%
brightness: 100%
effect: none
- delay: 1s
- script.execute: reset_led
- script.wait: reset_led
- lambda: |-
if (code == "wake-provider-missing" || code == "wake-engine-missing") {
id(use_wake_word).turn_off();
}
script:
- id: reset_led
then:
- if:
condition:
switch.is_on: use_wake_word
then:
- light.turn_on:
id: led_ring
blue: 100%
red: 0%
green: 0%
brightness: 30%
effect: none
else:
- light.turn_off: led_ring
switch:
- platform: gpio
id: pa_ctrl
pin: GPIO38
name: "${friendly_name} Speaker Mute"
restore_mode: ALWAYS_ON
- platform: template
name: Use wake word
id: use_wake_word
optimistic: true
restore_mode: RESTORE_DEFAULT_ON
entity_category: config
on_turn_on:
- lambda: id(voice_asst).set_use_wake_word(true);
- if:
condition:
not:
- voice_assistant.is_running
then:
- voice_assistant.start_continuous
- script.execute: reset_led
on_turn_off:
- voice_assistant.stop
- script.execute: reset_led
light:
- platform: esp32_rmt_led_strip
id: led_ring
name: "${friendly_name} Light"
pin: GPIO19
num_leds: 12
rmt_channel: 0
rgb_order: GRB
chipset: ws2812
default_transition_length: 0s
effects:
- pulse:
name: "Pulse"
transition_length: 0.5s
update_interval: 0.5s
- addressable_twinkle:
name: "Working"
twinkle_probability: 5%
progress_interval: 4ms
- addressable_color_wipe:
name: "Wakeword"
colors:
- red: 0%
green: 50%
blue: 0%
num_leds: 12
add_led_interval: 20ms
reverse: false
- addressable_color_wipe:
name: "Connecting"
colors:
- red: 60%
green: 60%
blue: 60%
num_leds: 12
- red: 60%
green: 60%
blue: 0%
num_leds: 12
add_led_interval: 100ms
reverse: true
binary_sensor:
- platform: template
name: "${friendly_name} Volume Up"
id: btn_volume_up
- platform: template
name: "${friendly_name} Volume Down"
id: btn_volume_down
- platform: template
name: "${friendly_name} Set"
id: btn_set
- platform: template
name: "${friendly_name} Play"
id: btn_play
- platform: template
name: "${friendly_name} Mode"
id: btn_mode
on_multi_click:
- timing:
- ON for at least 10s
then:
- button.press: factory_reset_btn
- platform: template
name: "${friendly_name} Record"
id: btn_record
on_press:
- voice_assistant.start:
- light.turn_on:
id: led_ring
brightness: 100%
effect: "Wakeword"
on_release:
- voice_assistant.stop:
- light.turn_off:
id: led_ring
sensor:
- id: button_adc
platform: adc
internal: true
pin: 8
attenuation: 11db
update_interval: 15ms
filters:
- median:
window_size: 5
send_every: 5
send_first_at: 1
- delta: 0.1
on_value_range:
- below: 0.55
then:
- binary_sensor.template.publish:
id: btn_volume_up
state: ON
- above: 0.65
below: 0.92
then:
- binary_sensor.template.publish:
id: btn_volume_down
state: ON
- above: 1.02
below: 1.33
then:
- binary_sensor.template.publish:
id: btn_set
state: ON
- above: 1.43
below: 1.77
then:
- binary_sensor.template.publish:
id: btn_play
state: ON
- above: 1.87
below: 2.15
then:
- binary_sensor.template.publish:
id: btn_mode
state: ON
- above: 2.25
below: 2.56
then:
- binary_sensor.template.publish:
id: btn_record
state: ON
- above: 2.8
then:
- binary_sensor.template.publish:
id: btn_volume_up
state: OFF
- binary_sensor.template.publish:
id: btn_volume_down
state: OFF
- binary_sensor.template.publish:
id: btn_set
state: OFF
- binary_sensor.template.publish:
id: btn_play
state: OFF
- binary_sensor.template.publish:
id: btn_mode
state: OFF
- binary_sensor.template.publish:
id: btn_record
state: OFF
Many people have already written about the fact that crackling and noise are primarily caused by interference in the wiring between the ESP and the sound module board. Because it is not just a relay or LED control wire, but a digital serial data bus operating at high frequencies. The best solution is to use no wires at all. I’m sure everyone has these things, they are usually bundled with the boards. I made such an assembly, I’m sharing the pictures. No noise when turned on, in operation, crystal clear sound in music and voice.
In my opinion, this is more reliable and compact.
The sound module board (amplifier) should be powered directly from 5 volts, the microphone board from 3.3 volts. I took out a separate type-c power connector and soldered it to the esp directly to the inputs of the 3.3 volt step-down converter (don’t mess it up). This gives the least amount of voltage dropout.
And yes, I confirm that microphones can be easily killed when soldering. In my case, flux got into the hole and I couldn’t get it out with alcohol and other cleaners.
p.s.: And another mounting option (this is not for a speaker, but as an adapter to a 5.1 system)
5 Likes
I have a question. If we use a Bluetooth speaker like the one in this build, why using audio output through amplifier (that may generate noises and headaches)? Can we use instead the bloototh connection (that is usually available in all ESP32 boards) to connect to the speaker itself for audio output? Hoping, by doing so, we can use the volume controls on the speaker without supplementary programming (I’m guessing), and remove the problematic amplifier from the build. Esp32 board will be used for wakeword, incoming audio through microphone and led lights. Can it be done?
I don’t believe this is currently possible. Most examples I’ve seen or used turn off Bluetooth when microwakeword is enabled. Someone in another thread said you can output audio to a media player, but it has to be defined in the ESPHome configuration and you may or may not have to allow the device to make HA service calls. Just be sure to enable it on the device after it’s added to HA. You would use one of the 2 services below
# In some trigger
on_...:
# Simple
- homeassistant.service:
service: notify.html5
data:
message: Button was pressed
# With templates and variables
- homeassistant.service:
service: notify.html5
data:
title: New Humidity
data_template:
message: The humidity is {{ my_variable }}%.
variables:
my_variable: |-
return id(my_sensor).state;
I have had some time to look at this. And the microphone part is not possible, unless you have a device with headset_head_unit profile. PulseAudio does not support handsfree_head_unit profile, which most Bluetooth speaker would have, like mine. I could use the audio output of the speaker with the add-on, but unfortunately I couldn’t get the microphone work through Bluetooth.
If PulseAudio would have ofono as well, then it could work, but I guess, that is a different issue with HAOS.