We propose to use the personalized speech synthesis and the neural language generator to synthesize content relevant personalized speech for rapid speaker adaptation. It has two distinct aspects: First, it relieves the general data sparsity issue in rapid adaptation via making use of additional synthesized personalized speech; Second, it circumvents the obstacle of the explicit labeling error in unsupervised adaptation by converting it to pseudo-supervised adaptation. In this setup, the labeling error is implicitly rendered as less damaging speech distortion in the personalized synthesized speech. This results in significant performance breakthrough in the rapid unsupervised speaker adaptation. We apply the proposed methodology to a speaker adaptation task in a state-of-art speech transcription system. With 1 minute (min) adaptation data, our proposed approach yields 9.19 % or 5.98 % relative word error rate (WER) reduction for the supervised and the unsupervised adaptation, comparing to the negligible gain when adapting only with 1 min original speech. With 10 min adaptation data, it yields 12.53 % or 7.89 % relative WER reduction, doubling the gain of the baseline adaptation. The proposed approach is particularly suitable for unsupervised adaptation.

Rapid unsupervised speaker adaptation in an E2E system posits us new challenges due to its end-to-end unified structure in addition to its intrinsic difficulty of data sparsity and imperfect label [1]. Previously we proposed utilizing the content relevant personalized speech synthesis for rapid speaker adaptation and achieved significant performance breakthrough in a hybrid system [2]. In this paper, we answer the following two questions: First, how to effectively perform rapid speaker adaptation in an RNN-T. Second, whether our previously proposed approach is still beneficial for the RNN-T and what are the modification and distinct observations. We apply the proposed methodology to a speaker adaptation task in a state-of-art presentation transcription RNN-T system. In the 1 min setup, it yields 11.58 % or 7.95 % relative word error rate (WER) reduction for the sup/unsup adaptation, comparing to the negligible gain when adapting with 1 min source speech. In the 10 min setup, it yields 15.71 % or 8.00 % relative WER reduction, doubling the gain of the source speech adaptation. We further apply various data filtering techniques and significantly bridge the gap between sup/unsup adaptation.

We explore character-level neural network models for inferring punctuation from text-only input. Punctuation inference is treated as a sequence tagging problem where the input is a sequence of un-punctuated characters, and the output is a corresponding sequence of punctuation tags. We experiment with six architectures, all of which use a long short-term memory (LSTM) network for sequence modeling. They differ in the way the context and lookahead for a given character is derived: from simple character embedding and delayed output to enable lookahead, to complex convolutional neural networks (CNN) to capture context. We demonstrate that the accuracy of proposed character-level models are competitive with the accuracy of a state-of-the-art word-level Conditional Random Field (CRF) baseline with carefully crafted features.

We present a stochastic finite-state model for segmenting Chinese text into dictionary entries and productively derived words, and providing pronunciations for these words; the method incorporates a class-based model in its treatment of personal names. We also evaluate the system's performance, taking into account the fact that people often do not agree on a single segmentation.

Modifications to a clustering algorithm in which objects are adaptively partitioned into clusters of equal size are described. The object migration automaton has the advantage of being conceptually simple and easy to implement. Unfortunately, the algorithm may exhibit slow convergence speed and in some cases may not converge at all. The algorithm is modified to provide remedies to these conditions. Through experimental results, the modifications are shown to yield a substantial speedup in convergence (while maintaining 100% accuracy), especially as the number of objects to be partitioned increases.>