Neural networks for Text-to-Speech evaluation

• URL: http://arxiv.org/abs/2604.08562v1
• Date: Tue, 17 Mar 2026 16:07:15 GMT
• Title: Neural networks for Text-to-Speech evaluation
• Abstract summary: Speech-to-Speech systems need to deliver human-perceived quality at scale.<n>Human subjective evaluation protocols such as Mean Opinion Score (MOS) and Side-by-Side (SBS) comparisons remain the de facto gold standards.<n>This study addresses these barriers by formulating, and implementing a suite of novel neural models.
• Score: 35.277885003866466
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Ensuring that Text-to-Speech (TTS) systems deliver human-perceived quality at scale is a central challenge for modern speech technologies. Human subjective evaluation protocols such as Mean Opinion Score (MOS) and Side-by-Side (SBS) comparisons remain the de facto gold standards, yet they are expensive, slow, and sensitive to pervasive assessor biases. This study addresses these barriers by formulating, and implementing a suite of novel neural models designed to approximate expert judgments in both relative (SBS) and absolute (MOS) settings. For relative assessment, we propose NeuralSBS, a HuBERT-backed model achieving 73.7% accuracy (on SOMOS dataset). For absolute assessment, we introduce enhancements to MOSNet using custom sequence-length batching, as well as WhisperBert, a multimodal stacking ensemble that combines Whisper audio features and BERT textual embeddings via weak learners. Our best MOS models achieve a Root Mean Square Error (RMSE) of ~0.40, significantly outperforming the human inter-rater RMSE baseline of 0.62. Furthermore, our ablation studies reveal that naively fusing text via cross-attention can degrade performance, highlighting the effectiveness of ensemble-based stacking over direct latent fusion. We additionally report negative results with SpeechLM-based architectures and zero-shot LLM evaluators (Qwen2-Audio, Gemini 2.5 flash preview), reinforcing the necessity of dedicated metric learning frameworks.

Related papers

• Evaluating Zero-Shot and One-Shot Adaptation of Small Language Models in Leader-Follower Interaction [1.3511057160494195]
  Leader-follower interaction is an important paradigm in human-robot interaction (HRI)<n>Small language models (SLMs) offer a potential alternative, but their effectiveness for role classification in HRI has not been systematically evaluated.
  arXiv  Detail & Related papers  (2026-02-26T18:20:26Z)
• SpeechQualityLLM: LLM-Based Multimodal Assessment of Speech Quality [2.1178416840822027]
  Speech quality assessment is central to telephony, VoIP, and streaming systems, where large volumes of degraded audio must be monitored and optimized at scale.<n>We introduce SpeechQualityLLM, a multimodal speech quality question-answering (QA) system that couples an audio encoder with a language model and is trained on the NISQA corpus using template-based question-answer pairs.<n>Our system is supervised to generate textual answers from which numeric predictions are parsed and evaluated with standard regression and ranking metrics.
  arXiv  Detail & Related papers  (2025-12-09T04:39:50Z)
• AHELM: A Holistic Evaluation of Audio-Language Models [78.20477815156484]
  multimodal audio-language models (ALMs) take interleaved audio and text as input and output text.<n>AHELM is a benchmark that aggregates various datasets -- including 2 new synthetic audio-text datasets called PARADE and CoRe-Bench.<n>We also standardize the prompts, inference parameters, and evaluation metrics to ensure equitable comparisons across models.
  arXiv  Detail & Related papers  (2025-08-29T07:40:39Z)
• SALF-MOS: Speaker Agnostic Latent Features Downsampled for MOS Prediction [1.8862680628828246]
  Evaluation of voice synthesis can be done using objective metrics or subjective metrics.<n>Speaker Agnostic Latent Features (SALF)-Mean Opinion Score (MOS) is a small-sized, end-to-end, highly generalized and scalable model for predicting MOS score on a scale of 5.<n>We use the sequences of convolutions and stack them to get the latent features of the audio samples to get the best results based on mean squared error (MSE), Linear Concordance Correlation coefficient (LCC), Spearman Rank Correlation Coefficient (SRCC) and Kendall Rank Correlation Coefficient (KTAU)
  arXiv  Detail & Related papers  (2025-06-02T10:45:40Z)
• $C^2$AV-TSE: Context and Confidence-aware Audio Visual Target Speaker Extraction [80.57232374640911]
  We propose a model-agnostic strategy called the Mask-And-Recover (MAR)<n>MAR integrates both inter- and intra-modality contextual correlations to enable global inference within extraction modules.<n>To better target challenging parts within each sample, we introduce a Fine-grained Confidence Score (FCS) model.
  arXiv  Detail & Related papers  (2025-04-01T13:01:30Z)
• Audio Large Language Models Can Be Descriptive Speech Quality Evaluators [46.765203628127345]
  We introduce the first natural language-based speech evaluation corpus, generated from authentic human ratings.<n>This corpus offers detailed analysis across multiple dimensions and identifies causes of quality degradation.<n>We propose an alignment approach with LLM distillation (ALLD) to guide the audio LLM in extracting relevant information from raw speech.
  arXiv  Detail & Related papers  (2025-01-27T22:47:51Z)
• CCATMos: Convolutional Context-aware Transformer Network for Non-intrusive Speech Quality Assessment [12.497279501767606]
  We propose a novel end-to-end model structure called Convolutional Context-Aware Transformer (CCAT) network to predict the mean opinion score (MOS) of human raters. We evaluate our model on three MOS-annotated datasets spanning multiple languages and distortion types and submit our results to the ConferencingSpeech 2022 Challenge.
  arXiv  Detail & Related papers  (2022-11-04T16:46:11Z)
• SOMOS: The Samsung Open MOS Dataset for the Evaluation of Neural Text-to-Speech Synthesis [50.236929707024245]
  The SOMOS dataset is the first large-scale mean opinion scores (MOS) dataset consisting of solely neural text-to-speech (TTS) samples. It consists of 20K synthetic utterances of the LJ Speech voice, a public domain speech dataset.
  arXiv  Detail & Related papers  (2022-04-06T18:45:20Z)
• LDNet: Unified Listener Dependent Modeling in MOS Prediction for Synthetic Speech [67.88748572167309]
  We present LDNet, a unified framework for mean opinion score (MOS) prediction. We propose two inference methods that provide more stable results and efficient computation.
  arXiv  Detail & Related papers  (2021-10-18T08:52:31Z)
• Towards a Competitive End-to-End Speech Recognition for CHiME-6 Dinner Party Transcription [73.66530509749305]
  In this paper, we argue that, even in difficult cases, some end-to-end approaches show performance close to the hybrid baseline. We experimentally compare and analyze CTC-Attention versus RNN-Transducer approaches along with RNN versus Transformer architectures. Our best end-to-end model based on RNN-Transducer, together with improved beam search, reaches quality by only 3.8% WER abs. worse than the LF-MMI TDNN-F CHiME-6 Challenge baseline.
  arXiv  Detail & Related papers  (2020-04-22T19:08:33Z)
• Joint Contextual Modeling for ASR Correction and Language Understanding [60.230013453699975]
  We propose multi-task neural approaches to perform contextual language correction on ASR outputs jointly with language understanding (LU) We show that the error rates of off the shelf ASR and following LU systems can be reduced significantly by 14% relative with joint models trained using small amounts of in-domain data.
  arXiv  Detail & Related papers  (2020-01-28T22:09:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.