ovos-audio

ovos-audio is the audio daemon for OpenVoiceOS. It handles text-to-speech (TTS) synthesis and sound playback, and it runs as a background service that connects to the OVOS MessageBus.

Install

pip install ovos-audio[extras]

The extras group installs this package with the default plugins. Without it, you must install and configure TTS modules yourself, as described below.

Configuration

ovos-audio reads its settings from mycroft.conf.

{

  // Text to Speech parameters
  "tts": {
    "module": "ovos-tts-plugin-server",
    "fallback_module": "ovos-tts-plugin-mimic",
    "ovos-tts-plugin-mimic": {
        "voice": "ap"
    }
  },

  // File locations of sounds to play for system events
  "sounds": {
    "start_listening": "snd/start_listening.wav",
    "end_listening": "snd/end_listening.wav",
    "acknowledge": "snd/acknowledge.mp3",
    "error": "snd/error.mp3"
  },

  // Mechanism used to play WAV audio files
  "play_wav_cmdline": "paplay %1 --stream-name=mycroft-voice",

  // Mechanism used to play MP3 audio files
  "play_mp3_cmdline": "mpg123 %1",

  // Mechanism used to play OGG audio files
  "play_ogg_cmdline": "ogg123 -q %1"
}

Persona Support

ovos-audio supports dialog-transformer plugins that rewrite generated speech to match a tone or persona. See the technical manual's dialog transformers page for the full transformer chain and priority rules.

For example, ovos-solver-plugin-openai-persona rewrites text before synthesis, based on a persona string. Sample personas:

"rewrite the text as if you were explaining it to a 5-year-old"
"rewrite the text as if it was an angry old man speaking"
"Add more 'dude'ness to it"

Example input and output with the "explain to a 5-year-old" persona:

Input: "Quantum mechanics is a branch of physics that describes the behavior of particles at the smallest scales."
Output: "Quantum mechanics is like a special kind of science that helps us understand really tiny things."

To enable the plugin, add this to mycroft.conf:

"dialog_transformers": {
    "ovos-dialog-transformer-openai-plugin": {
        "rewrite_prompt": "rewrite the text as if you were explaining it to a 5-year-old"
    }
}

Using Legacy AudioService

The legacy audio service handles audio playback through the old Mycroft API. See the implementation in mycroft-core and ovos-bus-client.

By default, OCP delegates to the legacy audio service when needed, so no action is required. If you disable OCP, this API becomes the sole media playback provider.

Note: once ovos-media is released, OCP and this API will be disabled by default and deprecated.

See docs/audio-service.md for the module reference, and the technical manual's OCP audio plugin page for how OCP relates to this subsystem.

{
    "enable_old_audioservice": true,
    "disable_ocp": true,
    "Audio": {
        "default-backend": "vlc",
        "backends": {
          "simple": {
            "type": "ovos_audio_simple",
            "active": true
          },
          "vlc": {
            "type": "ovos_vlc",
            "active": true
          }
        }
    }
  },
}

Legacy backend plugins:

ovos-vlc-plugin
ovos-audio-plugin-simple (no HTTPS support)
ovos-audio-plugin-mpv (recommended default)
ovos-media-plugin-chromecast
ovos-media-plugin-spotify

About OCP:

OCP was developed for mycroft-core under the legacy audio service system.
OCP is always the default audio plugin, unless "disable_ocp": true is set in the config.
OCP uses the legacy API internally to delegate playback when the GUI is unavailable, or when configured to do so.
OCP does not support old Mycroft CommonPlay skills. The "ocp_legacy" pipeline in ovos-core handles that instead.
ovos-media will fully replace OCP in ovos-audio 1.0.0.

Related Projects

ovos-media, the replacement for OCP and the legacy audio service
ovos-core, the assistant core that starts and manages ovos-audio
ovos-plugin-manager, plugin discovery for TTS, G2P, and audio backends
ovos-bus-client, the MessageBus client used to communicate with ovos-audio

See docs/index.md for a full architecture overview.

License

ovos-audio is licensed under the Apache License 2.0.

Metadata

Release files for ovos-audio 2.2.3a2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for ovos-audio 2.2.3a2
File	Size	Uploaded
ovos_audio-2.2.3a2.tar.gz	26.6 kB	Aug 31, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for ovos-audio 2.2.3a2
File	Interpreter	ABI	Platform	Reset
ovos_audio-2.2.3a2-py3-none-any.whl 31.6 kB Aug 31, 2026	Python 3	none	any	Details

Total release size: 58.2 kB

Release files / ovos_audio-2.2.3a2.tar.gz

Download URL	ovos_audio-2.2.3a2.tar.gz
Size	26.6 kB
Tags	Source
SHA-256 checksum How to use checksums	`e94f49eaf50b7ae710e9453e8b674bf0af612a0427b5d6a0ab8792c1a05dbce1`
BLAKE2b-256 checksum How to use checksums	`1f76d3c3820317cd0ff15706c0c78a6caee3e4b57b708f7d31bb9ee10efacec4`
Upload date	Aug 31, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/7.0.0 CPython/3.13.14`

Release files / ovos_audio-2.2.3a2-py3-none-any.whl

Download URL	ovos_audio-2.2.3a2-py3-none-any.whl
Size	31.6 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`1341792eb85f3e4424c098bd1eb9b3e189c3c72bd5b171ac624d6bb4ed4299f5`
BLAKE2b-256 checksum How to use checksums	`08ed989a44762bfa45b01505789ef6224c7762d5cae875048c1e27a7ba270b07`
Upload date	Aug 31, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/7.0.0 CPython/3.13.14`