VITS

by jaywalnut310 (research)

End-to-end TTS with adversarial training — the open-source workhorse.

TL;DR

End-to-end TTS with adversarial training — the open-source workhorse.

Best for engineers building custom TTS on a specific dataset where they control the training corpus. Pricing: free (MIT).

Category
Open source
License
—
Stars
—
Last push
—
Pricing
free (MIT)
Platforms
Linux

What it is

VITS (Variational Inference with adversarial learning for end-to-end TTS) is the 2021 architecture that underpins large chunks of subsequent TTS research, including Coqui's models. MIT-licensed reference implementation. Consent posture: training-time choice is operator's.

Best for: Engineers building custom TTS on a specific dataset where they control the training corpus.
Watch out for: Reference implementation — needs training from scratch on your own data for production use.

Install / use

git clone https://github.com/jaywalnut310/vits

Features

Speaker diarizationNo
Word-level timestampsNo
Streaming / real-timeNo
Languages supported1
HIPAA eligibleNo

VITS vs Whipscribe

FeatureVITSWhipscribe
CategoryOpen sourceTranscription APIs
Pricingfree (MIT)free beta
Speaker diarization—Yes
Word timestamps—Yes
Streaming—No
Languages199
PlatformsLinuxWeb, API, MCP

Alternatives to VITS

OpenAI Whisper
OpenAI

The reference open-source multilingual ASR model from OpenAI.

OSS · MIT ★ 98.1k
whisper.cpp
Georgi Gerganov

C/C++ port of Whisper — runs on anything, from a Raspberry Pi to Apple Silicon.

OSS · MIT ★ 48.8k
faster-whisper
SYSTRAN

4× faster than reference Whisper using CTranslate2 — production sweet spot.

OSS · MIT ★ 22.3k

Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.