Constrained Output Embeddings for End-to-End Code-Switching Speech Recognition with Only Monolingual Data
| Autor: | Haihua Xu, Van Tung Pham, Bin Ma, Eng Siong Chng, Yerbolat Khassanov, Zhiping Zeng, Chongjia Ni |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: |
FOS: Computer and information sciences
Training set Computer Science - Computation and Language Computer science Speech recognition InformationSystems_INFORMATIONSTORAGEANDRETRIEVAL Word error rate 010501 environmental sciences Code-switching 01 natural sciences Task (project management) End-to-end principle 0103 physical sciences Divergence (statistics) 010301 acoustics Computation and Language (cs.CL) 0105 earth and related environmental sciences |
| Zdroj: | INTERSPEECH |
| Popis: | The lack of code-switch training data is one of the major concerns in the development of end-to-end code-switching automatic speech recognition (ASR) models. In this work, we propose a method to train an improved end-to-end code-switching ASR using only monolingual data. Our method encourages the distributions of output token embeddings of monolingual languages to be similar, and hence, promotes the ASR model to easily code-switch between languages. Specifically, we propose to use Jensen-Shannon divergence and cosine distance based constraints. The former will enforce output embeddings of monolingual languages to possess similar distributions, while the later simply brings the centroids of two distributions to be close to each other. Experimental results demonstrate high effectiveness of the proposed method, yielding up to 4.5% absolute mixed error rate improvement on Mandarin-English code-switching ASR task. 5 pages, 3 figures, accepted to INTERSPEECH 2019 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|