Hierarchical Decoding for Discrete Speech Synthesis with Multi-Resolution Spoof Detection

2026-03-05

AI Summary

This work addresses the degradation in perceptual quality of discrete speech synthesis caused by token-level artifacts and distributional drift in neural codec language models. To mitigate these issues, the authors propose MSpoof-TTS, a training-free inference framework that introduces, for the first time, a multi-resolution token-level spoofing detection mechanism. Integrated within a hierarchical decoding process, this mechanism dynamically prunes low-quality token candidates and re-ranks hypotheses under discriminator guidance, enabling high-fidelity zero-shot speech generation. Notably, MSpoof-TTS operates without modifying model parameters, significantly enhancing the robustness and naturalness of synthesized speech while effectively suppressing locally inconsistent or unrealistic audio segments.

Technology Category

Natural Language Processing: SpeechMachine Learning: Large Multimodal Models (LMMs)Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applicationsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingWeb Mining and Content Analysis: Large pretrained models with web data

Abstract

Neural codec language models enable high-quality discrete speech synthesis, yet their inference remains vulnerable to token-level artifacts and distributional drift that degrade perceptual realism. Rather than relying on preference optimization or retraining, we propose MSpoof-TTS, a training-free inference framework that improves zero-shot synthesis through multi-resolution spoof guidance. We introduce a Multi-Resolution Token-based Spoof Detection framework that evaluates codec sequences at different temporal granularities to detect locally inconsistent or unnatural patterns. We then integrate the spoof detectors into a hierarchical decoding strategy, progressively pruning low-quality candidates and re-ranking hypotheses. This discriminator-guided generation enhances robustness without modifying model parameters. Experiments validate the effectiveness of our framework for robust and high-quality codec-based speech generation.

Problem

• discrete speech synthesis
• token-level artifacts
• distributional drift
• perceptual realism
• codec-based speech generation

Innovation

• multi-resolution spoof detection
• hierarchical decoding
• discrete speech synthesis
• training-free inference
• codec language models