# TensorSpeech/TensorFlowTTS

:stuck_out_tongue_closed_eyes: TensorFlowTTS: Real-Time State-of-the-art Speech Synthesis for Tensorflow 2 (supported including English, French, Korean, Chinese, German and Easy to adapt for other languages)

Repository: https://github.com/TensorSpeech/TensorFlowTTS
Canonical: https://ross.abutalabs.com/products/tensorflowtts
Homepage: https://tensorspeech.github.io/TensorFlowTTS/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: speech-synthesis, text-to-speech, tensorflow2, melgan, fastspeech, real-time, tts, vocoder, multi-speaker-tts, fastspeech2, multiband-melgan, tacotron2, parallel-wavegan, tflite, mobile-tts, zh-tts, chinese-tts, korea-tts, german-tts, japanese-tts
Last push: 2024-07-05T07:24:49+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2355, "days_push": 789, "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 3995, forks 797 (observed 2026-08-28T04:08:31.892243+00:00)

## What it is
TensorFlowTTS is a Python library providing real-time state-of-the-art text-to-speech architectures (Tacotron-2, FastSpeech/FastSpeech2, MelGAN, Multi-band MelGAN, HiFi-GAN) implemented in TensorFlow 2. It supports multiple languages including English, French, Korean, Chinese, German, and Japanese, and models can be quantized and converted to TFLite for mobile and embedded deployment.

## Use cases
- convert text to speech in python
- train a custom tts model
- deploy text-to-speech on mobile with tflite
- generate speech in chinese or korean
- run fastspeech2 with a melgan vocoder
- build a multilingual voice synthesis app
- real-time speech synthesis on embedded devices

## When to choose
- you need TTS models in the TensorFlow 2 ecosystem
- you want to deploy speech synthesis to mobile or embedded via TFLite
- you need pretrained models for several languages
- you want to experiment with FastSpeech2, Tacotron-2, or MelGAN-family vocoders

## When to avoid
- you prefer PyTorch-based TTS tooling
- you need actively developed cutting-edge features (development has slowed)
- you need production voice cloning or emotional TTS out of the box

## Facets
- artifact type: library
- maturity: maintenance
- function: tts, speech-recognition, deep-learning, machine-learning, audio-processing
- domain: speech-processing, machine-learning
- platform: python, cross-platform
- tags: text-to-speech, tensorflow2, vocoder, fastspeech2, tacotron2, melgan, hifi-gan, tflite, multilingual, real-time-synthesis, audio, natural-language-processing, mobile, gpu

## Member repositories
- TensorSpeech/TensorFlowTTS (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:31.892243+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 2026-08-29T18:24:11.411682+00:00, confidence not recorded.
  - readme: https://github.com/TensorSpeech/TensorFlowTTS (fetched 2026-08-28T04:08:31.892243+00:00, sha b3ddbdf272f1)
  - homepage: https://tensorspeech.github.io/TensorFlowTTS/ (fetched 2026-08-29T09:17:18.389039+00:00, sha 50750f495245)
  - registry_pypi: https://pypi.org/pypi/tensorflowtts/json (fetched 2026-08-29T09:17:18.391730+00:00, sha dea27c7eab2e)
- Data as of 2026-08-30T08:39:29.467469+00:00.