Cross-lingual Speaker Adaptation Research Articles

Using Eigenvoices and Nearest-Neighbors in HMM-Based Cross-Lingual Speaker Adaptation With Limited Data

Cross-lingual speaker adaptation for speech synthesis has many applications, such as use in speech-to-speech translation systems. Here, we focus on cross-lingual adaptation for statistical speech synthesis systems using limited adaptation data. To that end, we propose two eigenvoice adaptation approaches exploiting a bilingual Turkish-English speech database that we collected. In one approach, eigenvoice weights extracted using Turkish adaptation data and Turkish voice models are transformed into the eigenvoice weights for the English voice models using linear regression. Weighting the samples depending on the distance of reference speakers to target speakers during linear regression was found to improve the performance. Moreover, importance weighting the elements of the eigenvectors during regression further improved the performance. The second approach proposed here is speaker-specific state-mapping, which performed significantly better than the baseline state-mapping algorithm both in objective and subjective tests. Performance of the proposed state mapping algorithm was further improved when it was used with the intralingual eigenvoice approach instead of the linear-regression based algorithms used in the baseline system.

Analysis of unsupervised cross-lingual speaker adaptation for HMM-based speech synthesis using KLD-based transform mapping

In the EMIME project, we developed a mobile device that performs personalized speech-to-speech translation such that a user’s spoken input in one language is used to produce spoken output in another language, while continuing to sound like the user’s voice. We integrated two techniques into a single architecture: unsupervised adaptation for HMM-based TTS using word-based large-vocabulary continuous speech recognition, and cross-lingual speaker adaptation (CLSA) for HMM-based TTS. The CLSA is based on a state-level transform mapping learned using minimum Kullback–Leibler divergence between pairs of HMM states in the input and output languages. Thus, an unsupervised cross-lingual speaker adaptation system was developed. End-to-end speech-to-speech translation systems for four languages (English, Finnish, Mandarin, and Japanese) were constructed within this framework. In this paper, the English-to-Japanese adaptation is evaluated. Listening tests demonstrate that adapted voices sound more similar to a target speaker than average voices and that differences between supervised and unsupervised cross-lingual speaker adaptation are small. Calculating the KLD state-mapping on only the first 10 mel-cepstral coefficients leads to huge savings in computational costs, without any detrimental effect on the quality of the synthetic speech.

Current trends in multilingual speech processing

In this paper, we describe recent work at Idiap Research Institute in the domain of multilingual speech processing and provide some insights into emerging challenges for the research community. Multilingual speech processing has been a topic of ongoing interest to the research community for many years and the field is now receiving renewed interest owing to two strong driving forces. Firstly, technical advances in speech recognition and synthesis are posing new challenges and opportunities to researchers. For example, discriminative features are seeing wide application by the speech recognition community, but additional issues arise when using such features in a multilingual setting. Another example is the apparent convergence of speech recognition and speech synthesis technologies in the form of statistical parametric methodologies. This convergence enables the investigation of new approaches to unified modelling for automatic speech recognition and text-to-speech synthesis (TTS) as well as cross-lingual speaker adaptation for TTS. The second driving force is the impetus being provided by both government and industry for technologies to help break down domestic and international language barriers, these also being barriers to the expansion of policy and commerce. Speech-to-speech and speech-to-text translation are thus emerging as key technologies at the heart of which lies multilingual speech processing.

Personalising speech-to-speech translation: Unsupervised cross-lingual speaker adaptation for HMM-based speech synthesis

In this paper we present results of unsupervised cross-lingual speaker adaptation applied to text-to-speech synthesis. The application of our research is the personalisation of speech-to-speech translation in which we employ a HMM statistical framework for both speech recognition and synthesis. This framework provides a logical mechanism to adapt synthesised speech output to the voice of the user by way of speech recognition. In this work we present results of several different unsupervised and cross-lingual adaptation approaches as well as an end-to-end speaker adaptive speech-to-speech translation system. Our experiments show that we can successfully apply speaker adaptation in both unsupervised and cross-lingual scenarios and our proposed algorithms seem to generalise well for several language pairs. We also discuss important future directions including the need for better evaluation metrics.

Unsupervised Intralingual and Cross-Lingual Speaker Adaptation for HMM-Based Speech Synthesis Using Two-Pass Decision Tree Construction

Hidden Markov model (HMM)-based speech synthesis systems possess several advantages over concatenative synthesis systems. One such advantage is the relative ease with which HMM-based systems are adapted to speakers not present in the training dataset. Speaker adaptation methods used in the field of HMM-based automatic speech recognition (ASR) are adopted for this task. In the case of unsupervised speaker adaptation, previous work has used a supplementary set of acoustic models to estimate the transcription of the adaptation data. This paper first presents an approach to the unsupervised speaker adaptation task for HMM-based speech synthesis models which avoids the need for such supplementary acoustic models. This is achieved by defining a mapping between HMM-based synthesis models and ASR-style models, via a two-pass decision tree construction process. Second, it is shown that this mapping also enables unsupervised adaptation of HMM-based speech synthesis models without the need to perform linguistic analysis of the estimated transcription of the adaptation data. Third, this paper demonstrates how this technique lends itself to the task of unsupervised cross-lingual adaptation of HMM-based speech synthesis models, and explains the advantages of such an approach. Finally, listener evaluations reveal that the proposed unsupervised adaptation methods deliver performance approaching that of supervised adaptation.