Bert-VITS2

Bert-VITS2 is an open source text-to-speech project that combines the VITS2 backbone with multilingual BERT.

Bert-VITS2 is an open source text-to-speech project that utilizes a VITS2 backbone integrated with multilingual BERT. The core concept originates from MassTTS, and the repository includes references to projects such as vits, vits2pytorch, so-vits-svc, PaddleSpeech, emotional-vits, and fish-speech. It provides a webuipreprocess.py script for a quick guide on preprocessing and training. The project is intended for users who want to train speech models based on these frameworks. Note that the authors recommend the newer Fish-Speech project as an alternative.

Bert-VITS2 at a glance

Best for: Software engineer, Social media creator
Runs on: open source
Pricing: Free
Released: July 2023
Category: AI audio tool
Website: github.com/fishaudio/Bert-VITS2

Key features

VITS2 backbone architecture
Multilingual BERT integration
Web UI preprocessing script
Model training pipeline
Open source codebase

Use it for

Training text-to-speech models
Synthesizing speech with multilingual BERT
Running audio preprocessing scripts

Alternatives to Bert-VITS2

MOSS-TTS — Delivers zero-shot voice cloning and fine-grained pronunciation control.
WeNet — Achieves state-of-the-art results on public speech datasets.
GPT-SoVITS — It allows high-fidelity voice cloning with as little as one minute of data.
Fish Audio — Provides studio-grade voice cloning and text-to-speech with emotion control.
ESPnet — Provides reproducible recipes and hundreds of pretrained models for various speech tasks.
TTS WebUI — It offers a free web interface with over 30 AI models for audio generation.
RealtimeTTS — Provides low-latency speech synthesis with a broad engine matrix and multiple backends.
IndexTTS — Supports zero-shot voice cloning with multi-language emotion and speed control.
Speech-to-Speech — Fully modular, swappable components with local or hosted execution options.
AuK — It handles zero-shot TTS, editing, and enhancement via natural language.
Speaches — It acts as an Ollama equivalent for running local TTS and STT models.
Piper — Runs locally and offers fast neural text-to-speech synthesis.

All Bert-VITS2 alternatives