ANCHOR: Autoregressive Non-intrusive Chunk-Ordered Refinement for Joint Multi-Resolution Speech Quality Modeling
Not on the map. Why?
Abstract (arXiv)
While speech quality is typically assessed on complete utterances, streaming and generative systems require incremental estimation from partial audio. Existing predictors assume full context, degrading on prefix-constrained inputs. Extending ARECHO, we propose ANCHOR, reformulating incremental assessment as a multi-resolution autoregressive task. It models chunk- and utterance-level quality within a single decoder using dual-resolution tokens and a resolution-aware hierarchy for coarse-to-fine refinement. Experiments show substantial robustness under partial input, including a 48% PLCMOS error reduction on 2-second prefixes. Convergence analysis reveals a 4-6 s effective perceptual context horizon. A stress test further isolates structured extrapolation biases under localized corruption. Results demonstrate that hierarchical supervision improves incremental prediction and elucidates how perceptual quality accumulates over time.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 25 of its 39 references that are in the map. First posted on arXiv 8 Jun 2026.
Builds on
25 of 39 references.
References · 25
- 2021WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processingcited by 1,657
- 2019Common Voice: A Massively-Multilingual Speech Corpuscited by 1,384
- 2023Neural Codec Language Models are Zero-Shot Text to Speech Synthesizerscited by 752
- 2022AudioLM: a Language Modeling Approach to Audio Generationcited by 616
- 2017Tacotron: Towards End-to-End Speech Synthesiscited by 490
- 2022UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022cited by 390
- 2023Voicebox: Text-Guided Multilingual Universal Speech Generation at Scalecited by 379
- 2017AISHELL-1: An Open-Source Mandarin Speech Corpus and A Speech Recognition Baselinecited by 373
- 2021Layer-wise Analysis of a Self-supervised Speech Representation Modelcited by 298
- 2020DNSMOS: A Non-Intrusive Perceptual Objective Speech Quality metric to evaluate Noise Suppressorscited by 241
- 2018Efficient Neural Audio Synthesiscited by 219
- 2021NISQA: A Deep CNN-Self-Attention Model for Multidimensional Speech Quality Prediction with Crowdsourced Datasetscited by 192
13 more
- 2020Real Time Speech Enhancement in the Waveform Domaincited by 186
- 2023Speak, Read and Prompt: High-Fidelity Text-to-Speech with Minimal Supervisioncited by 169
- 2025Meta Audiobox Aesthetics: Unified Automatic Quality Assessment for Speech, Music, and Soundcited by 137
- 2019MOSNet: Deep Learning based Objective Assessment for Voice Conversioncited by 127
- 2020Emformer: Efficient Memory Transformer Based Acoustic Model For Low Latency Streaming Speech Recognitioncited by 113
- 2025Spark-TTS: An Efficient LLM-Based Text-to-Speech Model with Single-Stream Decoupled Speech Tokenscited by 113
- 2020Streaming automatic speech recognition with the transformer modelcited by 73
- 2022The VoiceMOS Challenge 2022cited by 57
- 2025Towards Frame-level Quality Predictions of Synthetic Speechcited by 4
- 2026UrgentMOS: Unified Multi-Metric and Preference Learning for Robust Speech Quality Assessmentcited by 4
- 2023NOMAD: Unsupervised Learning of Perceptual Embeddings for Speech Enhancement and Non-matching Reference Audio Quality Assessmentcited by 3
- 2025ARECHO: Autoregressive Evaluation via Chain-Based Hypothesis Optimization for Speech Multi-Metric Estimationcited by 2
- 2025Chunk Based Speech Pre-training with High Resolution Finite Scalar Quantizationcited by 2
Built on by
Citing papers per year: 2026: 1.
1 paper.