Meta-tts: Meta-learning For Few-shot Speaker Adaptive Text-to-speech

Abstract

Personalizing a speech synthesis system is a highly desired application, where the system can generate speech with the user's voice with rare enrolled recordings. There are two main approaches to build such a system in recent works: speaker adaptation and speaker encoding. On the one hand, speaker adaptation methods fine-tune a trained multi-speaker text-to-speech (TTS) model with few enrolled samples. However, they require at least thousands of fine-tuning steps for high-quality adaptation, making it hard to apply on devices. On the other hand, speaker encoding methods encode enrollment utterances into a speaker embedding. The trained TTS model can synthesize the user's speech conditioned on the corresponding speaker embedding. Nevertheless, the speaker encoder suffers from the generalization gap between the seen and unseen speakers. In this paper, we propose applying a meta-learning algorithm to the speaker adaptation method. More specifically, we use Model Agnostic Meta-Learning (

Stats

• citations49
• S2 citations—
• github stars0
• HF likes0
• heat score12.74
• arxiv keyhuang2021meta

Related papers

• Sample Efficient Adaptive Text-to-speech (2018)0.00
• Meta-voice: Fast Few-shot Style Transfer For Expressive Voice Cloning Using Meta Learning (2021)0.00
• The Universal Personalizer: Few-shot Dysarthric Speech Recognition Via Meta-learning (2025)0.00
• Learning To Adapt: A Meta-learning Approach For Speaker Adaptation (2018)9.76
• ASRRL-TTS: Agile Speaker Representation Reinforcement Learning For Text-to-speech Speaker Adaptation (2024)0.00
• Speaker-adaptive Neural Vocoders For Parametric Speech Synthesis Systems (2018)2.26
• Meta-learning Empowered Meta-face: Personalized Speaking Style Adaptation For Audio-driven 3D Talking Face Animation (2024)0.00
• Speaker Adaptive Training Using Model Agnostic Meta-learning (2019)9.92