Skip to content
Get started

USB Audio

The usb_audio component lets an ESP32 act as a USB host for USB Audio Class (UAC) devices such as USB sound cards, USB DACs, USB microphones and USB speakerphones. A connected device can be used as a USB Audio Speaker, a USB Audio Microphone, or both. Each usb_audio component supports one speaker and one microphone.

This component requires an ESP32 variant with a USB OTG peripheral (ESP32-S2, ESP32-S3, ESP32-S31, ESP32-P4 or ESP32-H4). It automatically loads the USB Host component, so it can be used alongside other USB host components such as USB UART.

WARNING

Audio and voice components consume a significant amount of resources (RAM, CPU) on the device.

Crashes are likely to occur if you include too many in the same configuration. In particular, Bluetooth/BLE components are known to cause issues when used in combination with audio and voice components.

# Example configuration entry
usb_audio:
  id: usb_audio_hub


speaker:
  - platform: usb_audio
    id: usb_speaker
    sample_rate: 48000
    bits_per_sample: 16bit
    num_channels: 2


microphone:
  - platform: usb_audio
    id: usb_mic
    sample_rate: 16000
    bits_per_sample: 16bit
    num_channels: 1

Configuration variables

Section titled “Configuration variables”
  • id (Optional, ID): The id to use for this component.

Stream format

Section titled “Stream format”

The speaker and microphone each stream in one fixed format, set by their sample_rate, bits_per_sample and num_channels options. The USB device must support that format. If it does not, the stream is not started and the formats the device does support are written to the log, so you can choose one of them. Devices that store samples padded to a larger size, such as 24-bit samples in 4 bytes, are not supported in that format.

On the ESP32-S2 and ESP32-S3, the USB controller has 800 bytes of usable FIFO memory for all USB packets, and the speaker and microphone streams get their share of it based on their formats. Formats with larger packets than ESP-IDF’s default split allows, including the speaker defaults of 48 kHz 16-bit stereo, make ESPHome size the FIFO itself, which needs ESP-IDF 5.5 or later (the default); on an older ESP-IDF, validation fails and says so. If the formats need more than the FIFO holds, for example a 96 kHz 32-bit stereo speaker and microphone together, validation fails; lower the sample rate, bits per sample or channels.

The device can be plugged in and removed while ESPHome is running. A speaker stops playback when the device is removed, and a microphone resumes streaming when the device is plugged in again.

See Also

Section titled “See Also”