# Attention-Constrained Inference for Robust Decoder-Only Text-to-Speech

arXiv: https://arxiv.org/abs/2404.19723 · HTML: https://mlsanity.com/paper/2404.19723/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Apr 2024.

## Built on by (most cited first, 6 of 6)

- [Recent Advances in Discrete Speech Tokens: A Review](https://mlsanity.com/paper/2502.06490/) (2025)
- [Whisper Has an Internal Word Aligner](https://mlsanity.com/paper/2509.09987/) (2025)
- [Eliminating stability hallucinations in llm-based tts models via attention guidance](https://mlsanity.com/paper/2509.19852/) (2025)
- [Gradient-Based Speech-to-Text Alignment for Any ASR Model: From CTC to Speech LLMs](https://mlsanity.com/paper/2607.06831/) (2026)
- [Read What You Hear: Reference-Free Hypotheses Evaluation with Acoustic Discrepancy](https://mlsanity.com/paper/2606.04680/) (2026)
- [Experience-Calibrated Contrastive Decoding for Mitigating Hallucinations in LM-Based Text-to-Speech](https://mlsanity.com/paper/2608.00722/) (2026)

## Abstract (arXiv)

Recent popular decoder-only text-to-speech models are known for their ability of generating natural-sounding speech. However, such models sometimes suffer from word skipping and repeating due to the lack of explicit monotonic alignment constraints. In this paper, we notice from the attention maps that some particular attention heads of the decoder-only model indicate the alignments between speech and text. We call the attention maps of those heads Alignment-Emerged Attention Maps (AEAMs). Based on this discovery, we propose a novel inference method without altering the training process, named Attention-Constrained Inference (ACI), to facilitate monotonic synthesis. It first identifies AEAMs using the Attention Sweeping algorithm and then applies constraining masks on AEAMs. Our experimental results on decoder-only TTS model VALL-E show that the WER of synthesized speech is reduced by up to 20.5% relatively with ACI while the naturalness and speaker similarity are comparable.