metavoiceio/metavoice-src
A 1.2B parameter text-to-speech model for human-like, expressive synthesis with zero-shot voice cloning.
Not currently ranked — collecting fresh signals.
star history
MetaVoice-1B is a foundational TTS model trained on 100K hours of speech using PyTorch. It supports emotional speech rhythm and tone in English, zero-shot voice cloning with 30-second reference audio, and cross-lingual voice cloning through fine-tuning. The model can synthesize speech of arbitrary length and is released under the Apache 2.0 license.
Frequently asked
- What is metavoiceio/metavoice-src?
- A 1.2B parameter text-to-speech model for human-like, expressive synthesis with zero-shot voice cloning.
- Is metavoice-src open source?
- Yes — metavoiceio/metavoice-src is open source, released under the Apache-2.0 license.
- What language is metavoice-src written in?
- metavoiceio/metavoice-src is primarily written in Python.
- How popular is metavoice-src?
- metavoiceio/metavoice-src has 4.2k stars on GitHub.
- Where can I find metavoice-src?
- metavoiceio/metavoice-src is on GitHub at https://github.com/metavoiceio/metavoice-src.