Joint Unsupervised and Supervised Training for Automatic Speech Recognition via Bilevel Optimization
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916102971326464 |
|---|---|
| author | Saif, A F M Cui, Xiaodong Shen, Han Lu, Songtao Kingsbury, Brian Chen, Tianyi |
| author_facet | Saif, A F M Cui, Xiaodong Shen, Han Lu, Songtao Kingsbury, Brian Chen, Tianyi |
| contents | In this paper, we present a novel bilevel optimization-based training approach to training acoustic models for automatic speech recognition (ASR) tasks that we term {bi-level joint unsupervised and supervised training (BL-JUST)}. {BL-JUST employs a lower and upper level optimization with an unsupervised loss and a supervised loss respectively, leveraging recent advances in penalty-based bilevel optimization to solve this challenging ASR problem with affordable complexity and rigorous convergence guarantees.} To evaluate BL-JUST, extensive experiments on the LibriSpeech and TED-LIUM v2 datasets have been conducted. BL-JUST achieves superior performance over the commonly used pre-training followed by fine-tuning strategy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_06980 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Joint Unsupervised and Supervised Training for Automatic Speech Recognition via Bilevel Optimization Saif, A F M Cui, Xiaodong Shen, Han Lu, Songtao Kingsbury, Brian Chen, Tianyi Computation and Language Machine Learning In this paper, we present a novel bilevel optimization-based training approach to training acoustic models for automatic speech recognition (ASR) tasks that we term {bi-level joint unsupervised and supervised training (BL-JUST)}. {BL-JUST employs a lower and upper level optimization with an unsupervised loss and a supervised loss respectively, leveraging recent advances in penalty-based bilevel optimization to solve this challenging ASR problem with affordable complexity and rigorous convergence guarantees.} To evaluate BL-JUST, extensive experiments on the LibriSpeech and TED-LIUM v2 datasets have been conducted. BL-JUST achieves superior performance over the commonly used pre-training followed by fine-tuning strategy. |
| title | Joint Unsupervised and Supervised Training for Automatic Speech Recognition via Bilevel Optimization |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2401.06980 |