mlsanityData to 14 Oct 2026

Analysis and Evaluation of Synthetic Data Generation in Speech Dysfluency Detection

arXiv 2505.22029 ↗First posted 28 May 2025eess.AS cs.AI cs.SDcited by 8

On the map: Synthetic data and self-improvement for LLMs → LLM synthetic data generation

Abstract (arXiv)

Speech dysfluency detection is crucial for clinical diagnosis and language assessment, but existing methods are limited by the scarcity of high-quality annotated data. Although recent advances in TTS model have enabled synthetic dysfluency generation, existing synthetic datasets suffer from unnatural prosody and limited contextual diversity. To address these limitations, we propose LLM-Dys -- the most comprehensive dysfluent speech corpus with LLM-enhanced dysfluency simulation. This dataset captures 11 dysfluency categories spanning both word and phoneme levels. Building upon this resource, we improve an end-to-end dysfluency detection framework. Experimental validation demonstrates state-of-the-art performance. All data, models, and code are open-sourced at https://github.com/Berkeley-Speech-Group/LLM-Dys.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Placed in LLM synthetic data generation, one of 13 directions in Synthetic data and self-improvement for LLMs. Builds on 25 of its 36 references that are in the map. First posted on arXiv 28 May 2025.

Builds on

25 of 36 references.

References · 25

  1. 2020Scaling Laws for Neural Language Modelscited by 6,586
  2. 2022Robust Speech Recognition via Large-Scale Weak Supervisioncited by 4,057
  3. 2019LibriTTS: A Corpus Derived from LibriSpeech for Text-to-Speechcited by 649
  4. 2021Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speechcited by 561
  5. 2024F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matchingcited by 361
  6. 2024E2 TTS: Embarrassingly Easy Fully Non-Autoregressive Zero-Shot TTScited by 152
  7. 2025Meta Audiobox Aesthetics: Unified Automatic Quality Assessment for Speech, Music, and Soundcited by 137
  8. 2021SEP-28k: A Dataset for Stuttering Event Detection From Podcasts With People Who Stuttercited by 52
  9. 2024CrisperWhisper: Accurate Timestamps on Verbatim Speech Transcriptionscited by 34
  10. 2016Disfluency Detection using a Bidirectional LSTMcited by 31
  11. 2022Detecting Dysfluencies in Stuttering Therapy Using wav2vec 2.0cited by 27
  12. 2023Unconstrained Dysfluency Modeling for Dysfluent Speech Transcription and Detectioncited by 21
13 more
  1. 2024Coding Speech through Vocal Tract Kinematicscited by 18
  2. 2024YOLO-Stutter: End-to-end Region-Wise Speech Dysfluency Detectioncited by 16
  3. 2024Towards Hierarchical Spoken Language Dysfluency Modelingcited by 15
  4. 2024SSDM: Scalable Speech Dysfluency Modelingcited by 13
  5. 2022Deep Neural Convolutive Matrix Factorization for Articulatory Representation Decompositioncited by 12
  6. 2023Deep Speech Synthesis from MRI-Based Articulatory Representationscited by 12
  7. 2022Enhancing ASR for Stuttered Speech with Limited Data Using Detect and Passcited by 11
  8. 2022Articulatory Representation Learning Via Joint Factor Analysis and Neural Matrix Factorizationcited by 11
  9. 2024Large Language Models for Dysfluency Detection in Stuttered Speechcited by 8
  10. 2024Stutter-Solver: End-to-end Multi-lingual Dysfluency Detectioncited by 8
  11. 2024Time and Tokens: Benchmarking End-to-End Speech Dysfluency Detectioncited by 5
  12. 2024Self-supervised Speech Models for Word-Level Stuttered Speech Detectioncited by 4
  13. 2024SSDM 2.0: Time-Accurate Speech Rich Transcription with Non-Fluenciescited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 3.

8 papers.

  1. 2025Seamless Dysfluent Speech Text Alignment for Disordered Speech Analysiscited by 4
  2. 2025K-Function: Joint Pronunciation Transcription and Feedback for Evaluating Kids Language Functioncited by 1
  3. 2025Revisiting Rule-Based Stuttering Detection: A Comprehensive Analysis of Interpretable Models for Clinical Applicationscited by 0
  4. 2025A Comparative Study of Controllability, Explainability, and Performance in Dysfluency Detection Modelscited by 0
  5. 2025Deploying UDM Series in Real-Life Stuttered Speech Applications: A Clinical Evaluation Frameworkcited by 0
  6. 2026Aligning Stuttered-Speech Research with End-User Needs: Scoping Review, Survey, and Guidelinescited by 0
  7. 2026Cripping AI: Reimagining AI Through Lived Disability Experiencescited by 0
  8. 2026What Counts as an Error? Dual-Reference Benchmarking for Atypical ASRcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Synthetic Data Generation for Phrase Break Prediction with Large Language Model
  2. 2025Artificial Conversations, Real Results: Fostering Language Detection with Synthetic Data
  3. 2025VocalBench-DF: A Benchmark for Evaluating Speech LLM Robustness to Disfluency
  4. 2025LLM-based Generative Error Correction for Rare Words with Synthetic Data and Phonetic Context
  5. 2024Beyond Oversmoothing: Evaluating DDPM and MSE for Scalable Speech Synthesis in ASR
  6. 2024Instruction Data Generation and Unsupervised Adaptation for Speech Language Models
  7. 2025Demystifying Synthetic Data in LLM Pre-training: A Systematic Study of Scaling Laws, Benefits, and Pitfalls
  8. 2024Hybrid Training Approaches for LLMs: Leveraging Real and Synthetic Data to Enhance Model Performance in Domain-Specific Applications