SSDM: Scalable Speech Dysfluency Modeling
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914963988152320 |
|---|---|
| author | Lian, Jiachen Zhou, Xuanru Ezzes, Zoe Vonk, Jet Morin, Brittany Baquirin, David Mille, Zachary Tempini, Maria Luisa Gorno Anumanchipalli, Gopala Krishna |
| author_facet | Lian, Jiachen Zhou, Xuanru Ezzes, Zoe Vonk, Jet Morin, Brittany Baquirin, David Mille, Zachary Tempini, Maria Luisa Gorno Anumanchipalli, Gopala Krishna |
| contents | Speech dysfluency modeling is the core module for spoken language learning, and speech therapy. However, there are three challenges. First, current state-of-the-art solutions\cite{lian2023unconstrained-udm, lian-anumanchipalli-2024-towards-hudm} suffer from poor scalability. Second, there is a lack of a large-scale dysfluency corpus. Third, there is not an effective learning framework. In this paper, we propose \textit{SSDM: Scalable Speech Dysfluency Modeling}, which (1) adopts articulatory gestures as scalable forced alignment; (2) introduces connectionist subsequence aligner (CSA) to achieve dysfluency alignment; (3) introduces a large-scale simulated dysfluency corpus called Libri-Dys; and (4) develops an end-to-end system by leveraging the power of large language models (LLMs). We expect SSDM to serve as a standard in the area of dysfluency modeling. Demo is available at \url{https://berkeley-speech-group.github.io/SSDM/}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_16221 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SSDM: Scalable Speech Dysfluency Modeling Lian, Jiachen Zhou, Xuanru Ezzes, Zoe Vonk, Jet Morin, Brittany Baquirin, David Mille, Zachary Tempini, Maria Luisa Gorno Anumanchipalli, Gopala Krishna Audio and Speech Processing Artificial Intelligence Computation and Language Sound Speech dysfluency modeling is the core module for spoken language learning, and speech therapy. However, there are three challenges. First, current state-of-the-art solutions\cite{lian2023unconstrained-udm, lian-anumanchipalli-2024-towards-hudm} suffer from poor scalability. Second, there is a lack of a large-scale dysfluency corpus. Third, there is not an effective learning framework. In this paper, we propose \textit{SSDM: Scalable Speech Dysfluency Modeling}, which (1) adopts articulatory gestures as scalable forced alignment; (2) introduces connectionist subsequence aligner (CSA) to achieve dysfluency alignment; (3) introduces a large-scale simulated dysfluency corpus called Libri-Dys; and (4) develops an end-to-end system by leveraging the power of large language models (LLMs). We expect SSDM to serve as a standard in the area of dysfluency modeling. Demo is available at \url{https://berkeley-speech-group.github.io/SSDM/}. |
| title | SSDM: Scalable Speech Dysfluency Modeling |
| topic | Audio and Speech Processing Artificial Intelligence Computation and Language Sound |
| url | https://arxiv.org/abs/2408.16221 |