GLM-TTS Technical Report

TL;DR

GLM-TTS is a high-fidelity, efficient, and controllable text-to-speech system that leverages a two-stage architecture and reinforcement learning to optimize speech quality and customization with limited training data.

Contribution

The paper introduces a novel two-stage TTS architecture with reinforcement learning and parameter-efficient customization for production-level speech synthesis.

Findings

Achieves state-of-the-art performance on open-source benchmarks

Utilizes only 100k hours of training data

Enables real-time, controllable speech synthesis

Abstract

This work proposes GLM-TTS, a production-level TTS system designed for efficiency, controllability, and high-fidelity speech generation. GLM-TTS follows a two-stage architecture, consisting of a text-to-token autoregressive model and a token-to-waveform diffusion model. With only 100k hours of training data, GLM-TTS achieves state-of-the-art performance on multiple open-source benchmarks. To meet production requirements, GLM-TTS improves speech quality through an optimized speech tokenizer with fundamental frequency constraints and a GRPO-based multi-reward reinforcement learning framework that jointly optimizes pronunciation, speaker similarity, and expressive prosody. In parallel, the system enables efficient and controllable deployment via parameter-efficient LoRA-based voice customization and a hybrid phoneme-text input scheme that provides precise pronunciation control. Our code is available at https://github.com/zai-org/GLM-TTS. Real-time speech synthesis demos are provided via Z.ai (audio.z.ai), the Zhipu Qingyan app/web (chatglm.cn).