Voice Cloning

Voice manipulation functionality for fmus-vox.

This module provides functionality for voice cloning, transformation, and enhancement.

fmus_vox.voice.clone_voice(reference_audio: str | Audio, text: str, output: str | None = None, model: str = 'yourtts', **kwargs) → Audio | None[source]

Clone a voice from reference audio and synthesize text with it.

This is a simple functional API for quick voice cloning. For more control, use the VoiceCloner class directly.

Parameters:

  • reference_audio – Reference audio to clone voice from (file path or Audio object)

  • text – Text to synthesize with the cloned voice

  • output – Path to save audio file (if None, returns Audio object)

  • model – Model to use for voice cloning (yourtts, etc.)

  • **kwargs – Additional model-specific parameters

Returns:

Audio object if output is None, otherwise None

Raises:

VoiceError – If voice cloning fails

Examples

>>> # Clone voice and play it
>>> audio = clone_voice("my_voice.wav", "Hello, this is my cloned voice")
>>> audio.play()

>>> # Clone voice and save to file
>>> clone_voice("my_voice.wav", "Hello, this is my cloned voice", output="cloned.wav")

VoiceCloner Class

class fmus_vox.voice.cloner.VoiceCloner(model: str = 'yourtts', **kwargs)[source]

Bases: object

Base class for voice cloning.

This class provides the common interface for all voice cloning models and handles voice management, cloning, and synthesis.

Parameters:

  • model – Name of the model to use (yourtts, etc.)

  • device – Computation device (cpu, cuda, auto)

  • **kwargs – Additional model-specific parameters

classmethod register_model(name: str, implementation: type) → None[source]

Register a model implementation.

Parameters:

  • name – Model name

  • implementation – Model implementation class

static __new__(cls, model: str = 'yourtts', **kwargs) → VoiceCloner[source]

Create a new VoiceCloner instance of the appropriate subclass.

Parameters:

  • model – Name of the model to use

  • **kwargs – Additional model-specific parameters

Returns:

VoiceCloner instance

Raises:

ModelError – If the model is not supported

__init__(model: str = 'yourtts', device: str | None = None, **kwargs)[source]

Initialize the voice cloner.

Parameters:

  • model – Name of the model to use

  • device – Computation device (cpu, cuda, auto)

  • **kwargs – Additional model-specific parameters

add_reference(**kwargs)
synthesize(**kwargs)
get_voice(voice_id: str) → Voice[source]

Get a voice by ID.

Parameters:

voice_id – ID of the voice to get

Returns:

Voice object

Raises:

VoiceError – If voice ID is invalid

list_voices() → List[Dict[str, Any]][source]

List all registered voices.

Returns:

List of voice dictionaries with id, name, and metadata

remove_voice(voice_id: str) → None[source]

Remove a voice.

Parameters:

voice_id – ID of the voice to remove

Raises:

VoiceError – If voice ID is invalid