Control over voice quality, e.g. breathy and tense voice, is important for speech synthesis applications. For example, transformations can be used to modify aspects of the voice related to speaker’s identity and to improve expressiveness. However, it is hard to modify voice characteristics of the synthetic speech, without degrading speech quality. State-of-the-art statistical speech synthesiser...

ARTIC (Artificial Talker in Czech) is a corpusbased text-to-speech (TTS) system that enables to synthesise an arbitrary text, mainly for the Czech language. Basically, two versions of ARTIC are available—a single unit instance system (also known as fixed-inventory synthesis) with the quality of resulting speech limited by the fixed inventory, and multiple unit instance system with the quality p...

Anatomical differences and habitual long-term features are components in the individual voice quality and are cues about the speakers identity for the listener. Terms for voice quality or phonation types for use in normal speech often come from studies of pathological speech (laryngeal setting). Therefore they may not be well suited to describe voice quality (especially the variations) of a nor...

Total laryngectomy, as a consequence of carcinoma of the larynx, results in loss of speech function. Cerebrovascular stroke is the leading cause of reduced speech production ability, and thereby communication difficulties. The case is presented of a 60-year-old male patient who suffered stroke five years after a total laryngectomy. Speech rehabilitation was hampered due to the depressive state ...

This thesis deals with the occupational voice. In modern society, there is an increasing demand for oral communication in many professions. The label ‘occupational voice’ refers to those occupations where voice is the major tool. Employees in these occupations often suffer from voice symptoms to varying extents. In this thesis, the telephone customer service advisors are the target group of pro...

To construct a natural singing voice synthesis system, it is important to adequately control acoustic parameters such as fundamental frequency (F0), phoneme duration, and spectrum information in the synthesis method, based on comparative analysis between spokenand singing-voices. This paper proposes a transformation from read speech into singing-voice using STRAIGHT. This method is composed of ...

In this study three different prominence and speech melody related effects on voice quality were studied using an inverse filtering based method. The hypothesis that prominence as a function of sentence and word stress is signaled with more pressed voice quality was tested. The results indicated discrepancies in the parameterization results, and the original hypothesis could not be confirmed. I...

CNGL, School of Computer Science and Informatics, University College Dublin Dublin, Ireland {eva.szekely|mohamed.abou-zleikha}@ucdconnect.ie, {joao.cabral|peter.cahill|julie.berndsen}@ucd.ie Abstract Audiobooks are a rich resource of large quantities of natural sounding, highly expressive speech. In our previous research we have shown that it is possible to detect different expressive voice sty...

The key factors helping people to understand the synthesized voices of text-to-speech system are the naturalness and the intelligibility. However, making more natural voices remains a difficult task because of the speech data’s scarcity. With data limited corpus, prosodic information such as tone, intonation, Part-of-Speech is added to ensure the quality of synthetic speech. In the paper, we in...