# fishaudio/Bert-VITS2

vits2 backbone with multilingual-bert

Repository: https://github.com/fishaudio/Bert-VITS2
Canonical: https://ross.abutalabs.com/products/bert-vits2
Language: Python
License: AGPL-3.0
License Family: copyleft
Topics: bert, bert-vits2, tts, vits, vits2, bert-vits, llm, fish, vocoder, fish-speech, agent
Last push: 2026-08-24T19:14:49+00:00

## Health v2 (maintenance only)
Score: 63/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 8, longevity 80
- inputs: {"age_days": 1131, "days_push": 9, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 8796, forks 1305 (observed 2026-08-28T04:10:25.795007+00:00)

## What it is
Bert-VITS2 is a text-to-speech model implementation combining the VITS2 architecture with multilingual BERT embeddings, written in Python. It supports training custom voice models and includes a web UI preprocessing pipeline.

## Use cases
- train a custom text-to-speech voice model
- generate speech in multiple languages with bert embeddings
- clone a voice from a dataset
- run a tts webui for inference
- fine-tune vits2 for chinese speech synthesis

## When to choose
- you want to train a custom VITS2-based TTS voice with multilingual BERT
- you need an open-source voice cloning pipeline in Python

## When to avoid
- you need a state-of-the-art, actively maintained TTS - the authors recommend fish-speech instead
- you need commercial-friendly licensing - it is AGPL-3.0

## Facets
- artifact type: library
- maturity: maintenance
- function: tts, machine-learning, deep-learning, llm-inference
- domain: speech-processing, deep-learning
- platform: python, cross-platform
- tags: vits2, bert, voice-cloning, speech-synthesis, vocoder, text-to-speech, natural-language-processing

## Member repositories
- fishaudio/Bert-VITS2 (main) score 63

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:25.795007+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:25:12.046446+00:00, confidence not recorded.
  - readme: https://github.com/fishaudio/Bert-VITS2 (fetched 2026-08-28T04:10:25.795007+00:00, sha dd74c654e269)
- Data as of 2026-08-30T08:39:29.467469+00:00.