voice-kit-sample-collection.yaml

substitutions:
  wake_word_collector_server_ip: "<HA SERVER IP ADDRESS>"

esphome:
  name: esphome-voice-kit
  friendly_name: ESPHome Voice Kit Sample Collector
  name_add_mac_suffix: true
  platformio_options:
    board_build.flash_mode: dio
  project:
    name: esphome.voice-kit
    version: dev

esp32:
  board: esp32-s3-devkitc-1
  variant: esp32s3
  flash_size: 8MB
  framework:
    type: esp-idf
    version: recommended
    sdkconfig_options:
      CONFIG_ESP32S3_DEFAULT_CPU_FREQ_240: "y"
      CONFIG_ESP32S3_DATA_CACHE_64KB: "y"
      CONFIG_ESP32S3_DATA_CACHE_LINE_64B: "y"
      CONFIG_ESP32S3_INSTRUCTION_CACHE_32KB: "y"
      CONFIG_SPIRAM_ALLOW_STACK_EXTERNAL_MEMORY: "y"
      CONFIG_FREERTOS_USE_TRACE_FACILITY: "y"
      CONFIG_FREERTOS_GENERATE_RUN_TIME_STATS: "y"

      CONFIG_SPIRAM_TRY_ALLOCATE_WIFI_LWIP: "y"

      # Settings based on https://github.com/espressif/esp-adf/issues/297#issuecomment-783811702
      CONFIG_ESP32_WIFI_STATIC_RX_BUFFER_NUM: "16"
      CONFIG_ESP32_WIFI_DYNAMIC_RX_BUFFER_NUM: "512"
      CONFIG_ESP32_WIFI_STATIC_TX_BUFFER: "y"
      CONFIG_ESP32_WIFI_TX_BUFFER_TYPE: "0"
      CONFIG_ESP32_WIFI_STATIC_TX_BUFFER_NUM: "8"
      CONFIG_ESP32_WIFI_CACHE_TX_BUFFER_NUM: "32"
      CONFIG_ESP32_WIFI_AMPDU_TX_ENABLED: "y"
      CONFIG_ESP32_WIFI_TX_BA_WIN: "16"
      CONFIG_ESP32_WIFI_AMPDU_RX_ENABLED: "y"
      CONFIG_ESP32_WIFI_RX_BA_WIN: "32"
      CONFIG_LWIP_MAX_ACTIVE_TCP: "16"
      CONFIG_LWIP_MAX_LISTENING_TCP: "16"
      CONFIG_TCP_MAXRTX: "12"
      CONFIG_TCP_SYNMAXRTX: "6"
      CONFIG_TCP_MSS: "1436"
      CONFIG_TCP_MSL: "60000"
      CONFIG_TCP_SND_BUF_DEFAULT: "5840"
      CONFIG_TCP_WND_DEFAULT: "65535"  # Adjusted from linked settings to avoid compilation error
      CONFIG_TCP_RECVMBOX_SIZE: "512"
      CONFIG_TCP_QUEUE_OOSEQ: "y"
      CONFIG_TCP_OVERSIZE_MSS: "y"
      CONFIG_LWIP_WND_SCALE: "y"
      CONFIG_TCP_RCV_SCALE: "3"
      CONFIG_LWIP_TCPIP_RECVMBOX_SIZE: "512"

      CONFIG_BT_ALLOCATION_FROM_SPIRAM_FIRST: "y"
      CONFIG_BT_BLE_DYNAMIC_ENV_MEMORY: "y"

wifi:
  ssid: "<WIFI SSID>"
  password: "<WIFI PASSWORD>"
  on_connect:
    - micro_wake_word.start:

logger:
  level: DEBUG
  logs:
    sensor: WARN  # avoids logging debug sensor updates

api:
ota:
  - platform: esphome
    id: ota_esphome


improv_serial:

esp32_improv:
  authorizer: center_button

i2c:
  sda: GPIO5
  scl: GPIO6

psram:
  mode: octal
  speed: 80MHz

binary_sensor:
  # Center Button. Used for many things (See on_multi_click)
  - platform: gpio
    id: center_button
    pin:
      number: GPIO0
      inverted: true

switch:
  # Remove before production
  - platform: template
    name: "Upload wake word sample"
    restore_mode: RESTORE_DEFAULT_OFF
    turn_on_action:
      - lambda: id(mww).set_upload_status(true);
    turn_off_action:
      - lambda: id(mww).set_upload_status(false);
    optimistic: true



light:
  # Hardware LED ring. Not used because remapping needed
  - platform: esp32_rmt_led_strip
    id: leds_internal
    pin: GPIO21
    rmt_channel: 1
    num_leds: 12
    rgb_order: GRB
    chipset: WS2812
    default_transition_length: 0ms
    power_supply: led_power

  # Voice Assistant LED ring. Remapping of the internal LED.
  # This light is not exposed. The device controls it
  - platform: partition
    id: voice_assistant_leds
    internal: true
    default_transition_length: 0ms
    segments:
      - id: leds_internal
        from: 7
        to: 11
      - id: leds_internal
        from: 0
        to: 6

  # User facing LED ring. Remapping of the internal LEDs.
  # Exposed to be used by the user.
  - platform: partition
    id: led_ring
    name: LED Ring
    entity_category: config
    icon: "mdi:circle-outline"
    default_transition_length: 0ms
    restore_mode: RESTORE_DEFAULT_OFF
    segments:
      - id: leds_internal
        from: 7
        to: 11
      - id: leds_internal
        from: 0
        to: 6

power_supply:
  - id: led_power
    pin: GPIO45

sensor:
  # To be removed before launch
  - platform: debug
    free:
      name: "Heap Free"
    block:
      name: "Max Block Free"
    loop_time:
      name: "Loop Time"
    psram:
      name: "PSRAM Free"



i2s_audio:
  - id: i2s_input
    i2s_lrclk_pin:
      number: GPIO14
    i2s_bclk_pin:
      number: GPIO13

microphone:
  - platform: nabu_microphone
    i2s_din_pin: GPIO15
    i2s_mode: secondary
    adc_type: external
    id: parent_microphone
    pdm: false
    sample_rate: 16000
    bits_per_sample: 32bit
    i2s_audio_id: i2s_input
    channel_0:
      id: asr_mic
      amplify_shift: 0
    channel_1:
      id: comm_mic
      amplify_shift: 2

external_components:
  - source:
      type: git
      url: https://github.com/esphome/voice-kit
      ref: 20240911-mww-sample-ring-buffer
    components: [micro_wake_word]
    refresh: 0s
  
  - source:
      type: git
      url: https://github.com/esphome/voice-kit
      ref: dev
    components:
      - audio_dac
      - microphone
      - nabu_microphone
      - voice_assistant
      - voice_kit
    refresh: 0s

voice_kit:
  reset_pin: GPIO4
  firmware:
    url: https://github.com/esphome/voice-kit-xmos-firmware/releases/download/v1.3.1/ffva_v1.3.1_upgrade.bin
    version: "1.3.1"
    md5: 964635c5bf125529dab14a2472a15401


micro_wake_word:
  id: mww
  upload_host: ${wake_word_collector_server_ip}
  models:
    - model: okay_nabu
  vad:
  microphone: comm_mic

debug:
  update_interval: 5s