REVIEW 7 cited by
Time and Tokens: Benchmarking End-to-End Speech Dysfluency Detection
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Speech dysfluency modeling is a task to detect dysfluencies in speech, such as repetition, block, insertion, replacement, and deletion. Most recent advancements treat this problem as a time-based object detection problem. In this work, we revisit this problem from a new perspective: tokenizing dysfluencies and modeling the detection problem as a token-based automatic speech recognition (ASR) problem. We propose rule-based speech and text dysfluency simulators and develop VCTK-token, and then develop a Whisper-like seq2seq architecture to build a new benchmark with decent performance. We also systematically compare our proposed token-based methods with time-based methods, and propose a unified benchmark to facilitate future research endeavors. We open-source these resources for the broader scientific community. The project page is available at https://rorizzz.github.io/
Forward citations
Cited by 7 Pith papers
-
LCS-CTC: Leveraging Soft Alignments to Enhance Phonetic Transcription Robustness
LCS-CTC, a phoneme recognizer trained with similarity-aware LCS alignment masks constraining CTC, outperforms vanilla CTC on all reported PER, WPER, boundary-loss, and articulatory metrics.
-
Towards Accurate Phonetic Error Detection Through Phoneme Similarity Modeling
Soft phoneme-similarity labels in multi-task training improve verbatim phoneme recognition and yield two new pronunciation error metrics.
-
Analysis and Evaluation of Synthetic Data Generation in Speech Dysfluency Detection
The paper introduces LLM-Dys, a 12,790-hour synthetic dysfluent speech corpus generated by LLM plus TTS, and claims state-of-the-art dysfluency detection with a Whisper-based transcriber.
-
Seamless Dysfluent Speech Text Alignment for Disordered Speech Analysis
Neural LCS uses learned phoneme and word similarity instead of exact matches to align dysfluent speech to intended text, and it outperforms DTW and Hard LCS on simulated benchmarks.
-
Dysfluent WFST: A Framework for Zero-Shot Speech Dysfluency Transcription and Detection
A training-free WFST decoder that uses the reference text to constrain phoneme decoding reports large gains in dysfluent speech transcription and detection, though the baselines lack the same text information.
-
Revisiting Rule-Based Stuttering Detection: A Comprehensive Analysis of Interpretable Models for Clinical Applications
A hand-coded rule system with rate-normalized thresholds is reported to reach F1 0.86 on UCLASS for stuttering detection, but the supporting evaluation is largely unreproducible.
-
Smooth Operators: LLMs Translating Imperfect Hints into Disfluency-Rich Transcripts
An 8B LLaMa decoder with a Conformer audio encoder generates disfluency tokens and timestamps, and works even when the text hints come from imperfect phoneme or word aligners.
Discussion (0). Continue with ORCID to comment.