skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


  • NSF Public Access
  • Search Results
  • Improving Mispronunciation Detection with Wav2vec2-based Momentum Pseudo-Labeling for Accentedness and Intelligibility Assessment
Title: Improving Mispronunciation Detection with Wav2vec2-based Momentum Pseudo-Labeling for Accentedness and Intelligibility Assessment
Current leading mispronunciation detection and diagnosis (MDD) systems achieve promising performance via end-to-end phoneme recognition. One challenge of such end-to-end solutions is the scarcity of human-annotated phonemes on natural L2 speech. In this work, we leverage unlabeled L2 speech via a pseudo-labeling (PL) procedure and extend the fine-tuning approach based on pre-trained self-supervised learning (SSL) models. Specifically, we use Wav2vec 2.0 as our SSL model, and fine-tune it using original labeled L2 speech samples plus the created pseudo-labeled L2 speech samples. Our pseudo labels are dynamic and are produced by an ensemble of the online model on-the-fly, which ensures that our model is robust to pseudo label noise. We show that fine-tuning with pseudo labels achieves a 5.35% phoneme error rate reduction and 2.48% MDD F1 score improvement over a labeled-samples-only finetuning baseline. The proposed PL method is also shown to outperform conventional offline PL methods. Compared to the state-of-the-art MDD systems, our MDD solution produces a more accurate and consistent phonetic error diagnosis. In addition, we conduct an open test on a separate UTD-4Accents dataset, where our system recognition outputs show a strong correlation with human perception, based on accentedness and intelligibility.  more » « less
Award ID(s):
2140469
PAR ID:
10358184
Author(s) / Creator(s):
; ; ; ;
Date Published:
Journal Name:
Interspeech
ISSN:
1990-9772
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; ; (, ISCA INTERSPEECH-2023)
    N/A (Ed.)
    Automatic pronunciation assessment (APA) plays an important role in providing feedback for self-directed language learners in computer-assisted pronunciation training (CAPT). Several mispronunciation detection and diagnosis (MDD) systems have achieved promising performance based on end-to-end phoneme recognition. However, assessing the intelligibility of second language (L2) remains a challenging problem. One issue is the lack of large-scale labeled speech data from non-native speakers. Additionally, relying only on one aspect (e.g., accuracy) at a phonetic level may not provide a sufficient assessment of pronunciation quality and L2 intelligibility. It is possible to leverage segmental/phonetic-level features such as goodness of pronunciation (GOP), however, feature granularity may cause a discrepancy in prosodic-level (suprasegmental) pronunciation assessment. In this study, Wav2vec 2.0-based MDD and Goodness Of Pronunciation feature-based Transformer are employed to characterize L2 intelligibility. Here, an L2 speech dataset, with human-annotated prosodic (suprasegmental) labels, is used for multi-granular and multi-aspect pronunciation assessment and identification of factors important for intelligibility in L2 English speech. The study provides a transformative comparative assessment of automated pronunciation scores versus the relationship between suprasegmental features and listener perceptions, which taken collectively can help support the development of instantaneous assessment tools and solutions for L2 learners. 
    more » « less
  2. ; ; ; (, ISCA INTERSPEECH-2023)
    N/A (Ed.)
    This study is focused on understanding and quantifying the change in phoneme and prosody information encoded in the Self-Supervised Learning (SSL) model, brought by an accent identification (AID) fine-tuning task. This problem is addressed based on model probing. Specifically, we conduct a systematic layer-wise analysis of the representations of the Transformer layers on a phoneme correlation task, and a novel word-level prosody prediction task. We compare the probing performance of the pre-trained and fine-tuned SSL models. Results show that the AID fine-tuning task steers the top 2 layers to learn richer phoneme and prosody representation. These changes share some similarities with the effects of fine-tuning with an Automatic Speech Recognition task. In addition, we observe strong accent-specific phoneme representations in layer 9. To sum up, this study provides insights into the understanding of SSL features and their interactions with fine-tuning tasks. 
    more » « less
  3. ; (, Interspeech 2022)
    Speech emotion recognition (SER) is a challenging task due to the limited availability of real-world labeled datasets. Since it is easier to find unlabeled data, the use of self-supervised learning (SSL) has become an attractive alternative. This study proposes new pre-text tasks for SSL to improve SER. While our target application is SER, the proposed pre-text tasks include audio-visual formulations, leveraging the relationship between acoustic and facial features. Our proposed approach introduces three new unimodal and multimodal pre-text tasks that are carefully designed to learn better representations for predicting emotional cues from speech. Task 1 predicts energy variations (high or low) from a speech sequence. Task 2 uses speech features to predict facial activation (high or low) based on facial landmark movements. Task 3 performs a multi-class emotion recognition task on emotional labels obtained from combinations of action units (AUs) detected across a video sequence. We pre-train a network with 60.92 hours of unlabeled data, fine-tuning the model for the downstream SER task. The results on the CREMA-D dataset show that the model pre-trained on the proposed domain-specific pre-text tasks significantly improves the precision (up to 5.1%), recall (up to 4.5%), and F1-scores (up to 4.9%) of our SER system. 
    more » « less
  4. ; ; ; (, Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers))
    Muresan, Smaranda; Nakov, Preslav; Villavicencio, Aline (Ed.)
    Phonemes are defined by their relationship to words: changing a phoneme changes the word. Learning a phoneme inventory with little supervision has been a longstanding challenge with important applications to under-resourced speech technology. In this paper, we bridge the gap between the linguistic and statistical definition of phonemes and propose a novel neural discrete representation learning model for self-supervised learning of phoneme inventory with raw speech and word labels. Under mild assumptions, we prove that the phoneme inventory learned by our approach converges to the true one with an exponentially low error rate. Moreover, in experiments on TIMIT and Mboshi benchmarks, our approach consistently learns a better phoneme-level representation and achieves a lower error rate in a zero-resource phoneme recognition task than previous state-of-the-art self-supervised representation learning algorithms. 
    more » « less
  5. ; ; ; ; (, IEEE)
    Recent studies have demonstrated the effectiveness of fine-tuning self-supervised speech representation models for speech emotion recognition (SER). However, applying SER in real-world environments remains challenging due to pervasive noise. Relying on low-accuracy predictions due to noisy speech can undermine the user’s trust. This paper proposes a unified self-supervised speech representation framework for enhanced speech emotion recognition designed to increase noise robustness in SER while generating enhanced speech. Our framework integrates speech enhancement (SE) and SER tasks, leveraging shared self-supervised learning (SSL)-derived features to improve emotion classification performance in noisy environments. This strategy encourages the SE module to enhance discriminative information for SER tasks. Additionally, we introduce a cascade unfrozen training strategy, where the SSL model is gradually unfrozen and fine-tuned alongside the SE and SER heads, ensuring training stability and preserving the generalizability of SSL representations. This approach demonstrates improvements in SER performance under unseen noisy conditions without compromising SE quality. When tested at a 0 dB signal-to-noise ratio (SNR) level, our proposed method outperforms the original baseline by 3.7% in F1-Macro and 2.7% in F1-Micro scores, where the differences are statistically significant. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email