In the last years, Deep Bidirectional Recurrent Neural Networks (DBRNN) and DBRNN with Long Short-Term Memory cells (DBLSTM) have outperformed the most accurate classifiers for confidence estimation in automatic speech recognition. At the same time, we have recently shown that speaker adaptation of confidence measures using DBLSTM yields significant improvements over non-adapted confidence measures. In accordance with these two recent contributions to the state of the art in confidence estimation, this paper presents a comprehensive study of speaker-adapted confidence measures using DBRNN and DBLSTM models. Firstly, we present new empirical evidences of the superiority of RNN-based confidence classifiers evaluated over a large speech corpus consisting of the English LibriSpeech and the Spanish poliMedia tasks. Secondly, we show new results on speaker-adapted confidence measures considering a multi-task framework in which RNN-based confidence classifiers trained with LibriSpeech are adapted to speakers of the TED-LIUM corpus. These experiments confirm that speaker-adapted confidence measures outperform their non-adapted counterparts. Lastly, we describe an unsupervised adaptation method of the acoustic DBLSTM model based on confidence measures which results in better automatic speech recognition performance.

Confidence estimation for automatic speech recognition has been very recently improved by using Recurrent Neural Networks (RNNs), and also by speaker adaptation (on the basis of Conditional Random Fields). In this work, we explore how to obtain further improvements by combining RNNs and speaker adaptation. In particular, we explore different speakerdependent and-independent data representations for Bidirectional Long Short Term Memory RNNs of various topologies. Empirical tests are reported on the LibriSpeech dataset showing that the best results are achieved by the proposed combination of RNNs and speaker adaptation.

Abstract. Over the past few years, online multimedia educational repositories have increased in number and popularity. The main aim of the transLectures project is to develop cost-effective solutions for producing accurate transcriptions and translations for large video lecture repositories, such as VideoLectures.NET or the Universitat Politècnica de València's repository, poliMedia. In this paper, we present the transLectures-UPV toolkit (TLK), which has been specifically designed to meet the requirements of the transLectures project, but can also be used as a conventional ASR toolkit. The main features of the current release include HMM training and decoding with speaker adaptation techniques (fCM-LLR). TLK has been tested on the VideoLectures.NET and poliMedia repositories, yielding very competitive results. TLK has been released under the permissive open source Apache License v2.0 and can be directly downloaded from the transLectures website.

Abstract. Online multimedia repositories are growing rapidly. However, language barriers are often difficult to overcome for many of the current and potential users. In this paper we describe a TTS Spanish system and we apply it to the synthesis of transcribed and translated video lectures. A statistical parametric speech synthesis system, in which the acoustic mapping is performed with either HMM-based or DNN-based acoustic models, has been developed. To the best of our knowledge, this is the first time that a DNN-based TTS system has been implemented for the synthesis of Spanish. A comparative objective evaluation between both models has been carried out. Our results show that DNN-based systems can reconstruct speech waveforms more accurately.

No abstract