Unit selection based text-to-speech systems can generally obtain high speech quality provided that the database is large enough. In embedded applications, the related memory requirements may be excessive and often the database needs to be both pruned and compressed to fit it into the available memory space. In this paper, we study the topic of database compression. In particular, the focus is on speaker-specific optimization of the quantizers used in the database compression. First, we introduce the simple concept of dynamic quantizer structures, facilitating the use of speaker-specific optimizations by enabling convenient run-time updates. Second, we show that significant memory savings can be obtained through speaker-specific retraining while perfectly maintaining the quantization accuracy, even when the memory required for the additional codebook data is taken into account. Thus, the proposed approach can be considered effective in reducing the conventionally large footprint of unit selection based text-to-speech systems. Index Terms: speech synthesis, unit selection, database compression
Most of the published voice conversion schemes do not consider detailed prosody modeling but only control the F0 level and range. However, the detailed prosody can also carry a significant amount of speaker identity related information. This paper introduces a new method for converting the prosody in voice conversion. A syllable-based prosodic codebook is used to predict the converted F0 using not only the source contour but also linguistic information and segmental durations. The selection of the most suitable target contour is carried out using a trained classification and regression tree. The F0 contours in the codebook are represented in a transformed domain which allows compression and fast comparison. The performance of the prosodic conversion is evaluated in a real voice conversion system. The results indicate a significant improvement in speaker identity and naturalness when compared to GMM (Gaussian mixture model) based pitch prediction approach.
This paper describes a new flexible delexicalization method based on glottal excited parametric speech synthesis scheme. The system utilizes inverse filtered glottal flow and all-pole modelling of the vocal tract. The method provides a possibility to retain and manipulate all relevant prosodic features of any kind of speech. Most importantly, the features include voice quality, which has not been properly modeled in earlier delexicalization methods. The functionality of the new method was tested in a prosodic tagging experiment aimed at providing word prominence data for a text-to-speech synthesis system. The experiment confirmed the usefulness of the method and further corroborated earlier evidence that linguistic factors influence the perception of prosodic prominence.
Most studies on Mandarin HTS (HMM-based text-to-speech system) have taken the initial/final as the basic acoustic units. It is, however, challenging to develop a multilingual HTS in a uniformed and consistent way since most of other languages use the phoneme as the basic phonetic unit. It becomes hard to apply cross-lingual adaptation which need map phonemes from each other, particularly in the case of unified ASR and HTS system due to the phoneme nature of most of the ASR systems. In this paper, we propose a phoneme based Mandarin HTS system, which has been systematically evaluated by comparing it with the initial/final system. The experimental results show that the use of phoneme as the acoustic unit for Mandarin HTS is a promising unified approach, thus enabling better and more uniform development with other languages while significantly reducing the number of acoustic units. The flat-start training scheme is also evaluated to show that the phoneme segmentation problem is solved without any performance degradation for phoneme based Mandarin HTS system. This performs an automatic approach without dependency with particular ASR system.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
