Copy page

MAmyshell-ai/OpenVoiceV2

OpenVoice

Runs OpenVoice tone-color transfer from a source utterance to an authorized target-speaker recording.

Text to speechVoiceHub-nativeopenvoice-v2-converterParameters: 32.8MLanguages: en, es +4Training: customLicense: Checkpoint-specific

Parameter metadata: Exact learned-parameter total for VoiceHub's audited native primary graph at the registered default selection; separately loaded auxiliary models are excluded.

Use this model Copy model ID Checkpoint

Resources

Paper Upstream GitHub VoiceHub source Open in Colab

Usage

Complete the VoiceHub installation once, then run this repository-authored example. Model pages intentionally contain no package-install command.

This example is maintained against VoiceHub's public API; it is not copied from an upstream demo or package README.

Model-specific path: Runs OpenVoice tone-color transfer from a source utterance to an authorized target-speaker recording.

Inputs and controls: base.wav must contain the utterance to convert; reference.wav supplies only the target tone color.

from pathlib import Path

from voicehub import AutoModelForTextToSpeech, TTSGenerationConfig

BASE_AUDIO = Path("base.wav")
REFERENCE_AUDIO = Path("reference.wav")
for audio_file in (BASE_AUDIO, REFERENCE_AUDIO):
    if not audio_file.is_file():
        raise FileNotFoundError(audio_file)

model = AutoModelForTextToSpeech.from_pretrained(
    'myshell-ai/OpenVoiceV2',
    model_type='openvoice',
    device="cuda",
    lazy_load=True,
)
output = model.generate(
    'VoiceHub keeps model integrations explicit and reproducible.',
    generation_config=TTSGenerationConfig(
        seed=42,
        output_file=Path("output.wav"),
    ),
    base_audio=str(BASE_AUDIO),
    speaker_audio_path=str(REFERENCE_AUDIO),
    tau=0.3,
)
print(output.file_path, output.sample_rate, output.metadata)

Use authorized recordings. Verify hardware needs and pin a revision in production.

Overview

openvoice is a VoiceHub text to speech integration. This page is generated from its registry contract. Open the openvoice Colab notebook.

Property	Value
Task	Text to speech
Architecture	openvoice-v2-converter
Runtime	VoiceHub-native
Languages	en, es, fr, zh, ja, ko
Capabilities	text-to-speech, voice-cloning, multilingual, fine-tuning, safetensors, voicehub-native, native-runtime, paired-waveform-training, explicit-base-waveform
Reusable components	wavmark
Normalized output	TTSOutput

Language support

Supported language abbreviations

en, es, fr, zh, ja, ko

Paper and GitHub

• Paper: OpenVoice: Versatile Instant Voice Cloning
• Upstream GitHub: OpenVoice
• VoiceHub source: VoiceHub model implementation

Configuration

Load configuration without constructing the model:

from voicehub import AutoConfig

config = AutoConfig.for_model('openvoice')
print(config.model_type)

Property	Value
Canonical model type	openvoice
Configuration class	OpenVoiceConfig
Architecture class	OpenVoiceForTextToSpeech

Processing

Create the registered processor without allocating model weights:

from voicehub import AutoProcessor

processor = AutoProcessor.from_pretrained(
    'myshell-ai/OpenVoiceV2',
    model_type='openvoice',
)
print(type(processor).__name__)

Inference

The Usage example returns TTSOutput through AutoModelForTextToSpeech.

Input and output contract

Property	Value
Readiness	integrated-raw
Data architecture	vits
Sample rate	22,050 Hz
Contract getter	get_tts_dataset_spec('openvoice')

Variant	Required fields	One of	Boundary	Other rules
paired-waveforms	source_audio, target_audio	—	Source	—
paired-waveform-aliases	audio, target_waveform	—	Source	—

VITS/GAN text, waveform, spectrogram, and adversarial data. See the data workflow.

Training and optimization

Use available_optimization_passes() to discover reversible public passes. Unsupported runtime or hardware fails closed before mutation.

Training contract

Property	Value
Support	custom
Family	vits
Recipe	single-phase
Default phase	generator
Training checkpoint	myshell-ai/OpenVoiceV2
Native training graph	yes

Phase	Kind	Components	Required inputs	Loss keys
generator	generator	model.enc_q, model.flow, model.dec, model.ref_enc	source_spectrogram, source_lengths, target_waveform, target_lengths	loss

This profile uses model-specific phases; inspect and honor each phase boundary. Call model.validate_training_support() first, then follow the training workflow.

Checkpoints, provenance, license, and limitations

Property	Value
Default checkpoint	myshell-ai/OpenVoiceV2
Hugging Face ID	myshell-ai/OpenVoiceV2 Repository availability verified through the Hugging Face model API on 2026-08-11; pin a revision before production use.
Checkpoint status	Registry default; pin an immutable revision for production and reproducible evidence
Optional dependency extra	Core package
Hardware and runtime	Usage selects cuda; verify checkpoint-specific requirements
Real-checkpoint evidence	Release evidence; a registry default alone is not execution evidence
Implementation	voicehub.models.openvoice.modeling_openvoice.OpenVoiceForTextToSpeech
Configuration	voicehub.models.openvoice.configuration_openvoice.OpenVoiceConfig
Source provenance	voicehub/models/openvoice/source/SOURCE.json
License	Checkpoint-specific

No VoiceHub-specific license override is registered. Verify the checkpoint and upstream source terms before use.

Confirm the checkpoint revision, access terms, provenance, and license.

Limitations

• No integration-specific checkpoint limitation is registered. Verify the selected checkpoint revision and its documented runtime requirements.
• Validate memory, precision, and optional dependencies on the target system.
• Public optimizations fail closed when the runtime or hardware cannot satisfy their validation contract; an unavailable pass is not reported as applied.
• Contract tests do not replace the linked released-checkpoint evidence.

Public API

Use the stable configuration, processor, and task-model facades below.

Configuration

OpenVoiceConfig

View source

OpenVoiceConfig(**config_kwargs)

Parameters

• **config_kwargs — Configuration fields validated by OpenVoiceConfig.

Model

OpenVoiceForTextToSpeech

View source

AutoModelForTextToSpeech.from_pretrained(
    pretrained_model_name_or_path,
    *,
    model_type='openvoice',
    config=None,
    **model_kwargs,
)

Parameters

• pretrained_model_name_or_path — Hub ID or compatible local directory.
• model_type — Canonical model type; use 'openvoice'.
• config — Optional preloaded OpenVoiceConfig instance.
• **model_kwargs — Model-specific loading arguments.

from voicehub import get_model_spec

spec = get_model_spec('openvoice')
print(spec.display_name, spec.task.value)

Purpose	Public object
Discover	get_model_spec('openvoice')
Load and run	AutoModelForTextToSpeech
Configure	OpenVoiceConfig
Process	AutoProcessor
Model implementation	OpenVoiceForTextToSpeech
Normalized output	TTSOutput
Training contract	get_training_spec('openvoice')
Optimization lifecycle	available_optimization_passes, apply_optimization_plan, optimization_manifest, restore_optimization_plan

See all model guides, inference, and the training matrix.