Multilingual Training and Cross-lingual Adaptation on CTC-based Acoustic Model

Abstract

Multilingual models for Automatic Speech Recognition (ASR) are attractive as they have been shown to benefit from more training data, and better lend themselves to adaptation to under-resourced languages. However, initialisation from monolingual context-dependent models leads to an explosion of context-dependent states. Connectionist Temporal Classification (CTC) is a potential solution to this as it performs well with monophone labels. We investigate multilingual CTC in the context of adaptation and regularisation techniques that have been shown to be beneficial in more conventional contexts. The multilingual model is trained to model a universal International Phonetic Alphabet (IPA)-based phone set using the CTC loss function. Learning Hidden Unit Contribution (LHUC) is investigated to perform language adaptive training. In addition, dropout during cross-lingual adaptation is also studied and tested in order to mitigate the overfitting problem. Experiments show that the performance of the universal phoneme-based CTC system can be improved by applying LHUC and it is extensible to new phonemes during cross-lingual adaptation. Updating all the parameters shows consistent improvement on limited data. Applying dropout during adaptation can further improve the system and achieve competitive performance with Deep Neural Network / Hidden Markov Model (DNN/HMM) systems on limited data.

References in corpus (5)

• The Microsoft 2016 Conversational Speech Recognition System
• Optimal Hyperparameters for Deep LSTM-Networks for Sequence Labeling Tasks
• Recurrent Dropout without Memory Loss
• Improving LSTM-CTC based ASR performance in domains with limited training data
• A Broadcast News Corpus for Evaluation and Tuning of German LVCSR Systems

Cited by in corpus (7)

• Cross-Language Transfer Learning, Continuous Learning, and Domain Adaptation for End-to-End Automatic Speech Recognition
• DARTS-ASR: Differentiable Architecture Search for Multilingual Speech Recognition and Adaptation
• Large-Scale Multilingual Speech Recognition with a Streaming End-to-End Model
• Adversarial Meta Sampling for Multilingual Low-Resource Speech Recognition
• Autosegmental Neural Nets: Should Phones and Tones be Synchronous or Asynchronous?
• Multilingual and crosslingual speech recognition using phonological-vector based phone embeddings
• Leveraging Text Data Using Hybrid Transformer-LSTM Based End-to-End ASR in Transfer Learning