Cant play on EspHome Device

[mvoss96]

Hi, i have configured Music assistant and can play on my existing commercial smart speakers, however i cant play on my diy esphome speaker via the HA MediaPlayers Integration. If i play a mp3 file or radio from home assistant everything works but if i try to play anything from music assistant i get no sound and errors in the esphome log:

Im using an ESP32S3 N8R2

44][D][api.connection:2229]: Home Assistant 2026.3.0 (192.168.2.51): connected
[20:21:44][D][binary_sensor:044]: 'Device Status' >> ON
[20:21:50][D][media_player:085]: 'Speaker Media Player' - Setting
[20:21:50][D][media_player:092]:   Media URL: http://192.168.2.51:8097/flow/UxC68SbT/media_player.web_radio_speaker_media_player/74a25fda512d451392d89d9455e99e7e.flac
[20:21:50][D][speaker_media_player:410]: State changed to PLAYING
[20:21:50][E][speaker_media_player.pipeline:112]: Media reader encountered an error: ESP_FAIL
[20:21:50][E][speaker_media_player:319]: The media pipeline's file reader encountered an error.
[20:21:50][E][speaker_media_player.pipeline:112]: Media reader encountered an error: ESP_FAIL
[20:21:50][E][speaker_media_player:319]: The media pipeline's file reader encountered an error.
[20:21:50][E][speaker_media_player.pipeline:112]: Media reader encountered an error: ESP_FAIL
[20:21:50][E][speaker_media_player:319]: The media pipeline's file reader encountered an error.
[20:21:50][E][speaker_media_player.pipeline:112]: Media reader encountered an error: ESP_FAIL
[20:21:50][E][speaker_media_player:319]: The media pipeline's file reader encountered an error.

my esphome config:

esphome:
  name: web-radio
  friendly_name: Web Radio

esp32:
  variant: ESP32S3
  flash_size: 8MB
  cpu_frequency: 240MHz
  framework:
    type: esp-idf

#wifi setup ....

########################################################################

psram:
  mode: quad
  speed: 80MHz
  ignore_not_found: false

i2s_audio:
  id: i2s_output
  i2s_lrclk_pin: GPIO05
  i2s_bclk_pin: GPIO04

speaker:
   # Hardware speaker output
  - platform: i2s_audio
    sample_rate: 48000
    id: i2s_speaker
    i2s_dout_pin: GPIO06
    bits_per_sample: 16bit
    i2s_audio_id: i2s_output
    dac_type: external
    channel: mono
    timeout: never
    buffer_duration: 100ms

  # Virtual speakers to combine the announcement and media streams together into one output
  - platform: mixer
    id: spk_mixer
    output_speaker: i2s_speaker
    num_channels: 1
    source_speakers:
      - id: announcement_spk_mixer_input
      - id: media_spk_mixer_input

  # Virtual speakers to resample each pipelines' audio, if necessary
  - platform: resampler
    id: media_spk_resampling_input
    output_speaker: media_spk_mixer_input
    sample_rate: 48000
    bits_per_sample: 16
  - platform: resampler
    id: announcement_spk_resampling_input
    output_speaker: announcement_spk_mixer_input
    sample_rate: 48000
    bits_per_sample: 16

media_player:
  - platform: speaker
    name: "Speaker Media Player"
    id: speaker_media_player
    announcement_pipeline:
      speaker: announcement_spk_resampling_input
      #format: MP3  
      num_channels: 1  
    media_pipeline: 
      speaker: media_spk_resampling_input
      #format: MP3
      num_channels: 1

[mitrokun]

Take a look at this configuration to see how you can implement a modern audio pipeline with sendspin support. You can ignore everything related to the voice assistant, but be sure to use external_components.

If you have any difficulties, I can share my code for S3 Zero (N4R2)