# zai-org/GLM-TTS

GLM-TTS: Controllable & Emotion-Expressive Zero-shot TTS with Multi-Reward Reinforcement Learning

Repository: https://github.com/zai-org/GLM-TTS
Canonical: https://ross.abutalabs.com/products/glm-tts
Homepage: https://audio.z.ai
Language: Python
License: Apache-2.0
License Family: permissive
Topics: edge-computing, llm, tts
Last push: 2026-04-10T08:50:23+00:00

## Health v2 (maintenance only)
Score: 50/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 76, release rhythm 35, longevity 19
- inputs: {"age_days": 270, "days_push": 145, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1055, forks 136 (observed 2026-08-28T04:03:24.287223+00:00)

## What it is
GLM-TTS is a Python-based text-to-speech synthesis system built on large language models, using a two-stage LLM plus Flow model architecture. It supports zero-shot voice cloning from short audio prompts, emotion-expressive speech via multi-reward reinforcement learning, and real-time streaming inference.

## Use cases
- clone a voice from a few seconds of audio
- generate emotional expressive speech from text
- stream tts audio in real time for interactive apps
- synthesize chinese and mixed chinese-english speech
- run text-to-speech locally on gpu or npu
- build a voice assistant with custom speaker voice

## When to choose
- you need zero-shot voice cloning with only 3-10 seconds of reference audio
- you want controllable emotion and prosody in synthesized speech
- you need streaming tts for real-time or interactive applications
- your primary language is Chinese with some English mixed in
- you want an open-source Apache-2.0 TTS model you can self-host

## When to avoid
- you need production-grade multilingual support beyond Chinese and English
- you have no GPU or NPU available for inference
- you need a lightweight CPU-only TTS solution
- you require a mature stable API with long-term support guarantees
- you need non-Python integration without wrapping the model

## Facets
- artifact type: library
- maturity: active
- function: tts, speech-recognition, llm-inference, machine-learning, deep-learning
- domain: speech-processing, artificial-intelligence, large-language-models
- platform: python, cross-platform
- tags: zero-shot-voice-cloning, text-to-speech, reinforcement-learning, streaming-inference, emotion-control, voice-cloning, flow-model, chinese, npu-support, natural-language-processing, gpu, linux

## Member repositories
- zai-org/GLM-TTS (main) score 50

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:24.287223+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 2026-08-30T06:58:42.151017+00:00, confidence not recorded.
  - readme: https://github.com/zai-org/GLM-TTS (fetched 2026-08-28T04:03:24.287223+00:00, sha de3037bf2b18)
  - homepage: https://audio.z.ai (fetched 2026-08-29T13:00:32.792178+00:00, sha 579779ec4a4c)
- Data as of 2026-08-30T08:39:29.467469+00:00.