tts

Kokoro TTS service implementation using kokoro-onnx.

pipecat.services.kokoro.tts.language_to_kokoro_language(language: Language) → str

Convert a Language enum to kokoro-onnx language code.

Parameters:

language – The Language enum value to convert.

Returns:

The corresponding kokoro-onnx locale code.

class pipecat.services.kokoro.tts.KokoroTTSSettings(model: str | None | _NotGiven = <factory>, extra: dict[str, Any]=<factory>, voice: str | None | _NotGiven = <factory>, language: Language | str | None | _NotGiven = <factory>)

Bases: TTSSettings

Settings for KokoroTTSService.

class pipecat.services.kokoro.tts.KokoroTTSService(*, voice_id: str | None = None, model_path: str | None = None, voices_path: str | None = None, params: InputParams | None = None, settings: KokoroTTSSettings | None = None, **kwargs)

Bases: TTSService

Kokoro TTS service implementation.

Provides local text-to-speech synthesis using kokoro-onnx. Automatically downloads model files on first use.

Settings

alias of KokoroTTSSettings

class InputParams(*, language: Language = Language.EN)

Bases: BaseModel

Input parameters for Kokoro TTS configuration.

Deprecated since version 0.0.105: Use KokoroTTSService.Settings directly via the settings parameter instead.

Parameters:

language – Language to use for synthesis.

language: Language

__init__(*, voice_id: str | None = None, model_path: str | None = None, voices_path: str | None = None, params: InputParams | None = None, settings: KokoroTTSSettings | None = None, **kwargs)

Initialize the Kokoro TTS service.

Parameters:

• voice_id –

  Voice identifier to use for synthesis.

  Deprecated since version 0.0.105: Use settings=KokoroTTSService.Settings(voice=...) instead.

• model_path – Path to the kokoro ONNX model file. Defaults to auto-downloaded file.

• voices_path – Path to the voices binary file. Defaults to auto-downloaded file.

• params –

  Configuration parameters for synthesis.

  Deprecated since version 0.0.105: Use settings=KokoroTTSService.Settings(...) instead.

• settings – Runtime-updatable settings. When provided alongside deprecated parameters, settings values take precedence.

• **kwargs – Additional arguments passed to parent TTSService.

can_generate_metrics() → bool

Indicate that this service supports TTFB and usage metrics.

language_to_service_language(language: Language) → str

Convert a Language enum to kokoro-onnx language format.

Parameters:

language – The language to convert.

Returns:

The kokoro-onnx language code.

async run_tts(text: str, context_id: str) → AsyncGenerator[Frame, None]

Synthesize speech from text using kokoro-onnx.

Uses the async streaming API to generate audio frames.

Parameters:

• text – The text to synthesize.

• context_id – Unique identifier for this TTS context.