Generalized source-filter structures for speech synthesis
نویسندگان
چکیده
In this paper we discuss various digital filter principles as models for synthetic speech generation. Warped linear prediction (WLP) and frequency-warped filters have been introduced earlier as a method to reduce the filter order in high-quality wideband speech synthesis. In addition to analyzing WLP and frequency-warped filters we introduce new related structures and techniques for arbitrary frequency resolution allocation. Kautz filters can be considered as generalized structures for pole-zero modeling. This study focuses on residual-excited synthesis and diphone-oriented reconstruction of speech signals. Control strategies for text-to-speech synthesis are discussed briefly.
منابع مشابه
Voice synthesis usingthe generalized pressure-Controlled valve
Vowel production in human speech depends both on the vocal tract shape, primarily establishing formant frequencies in the speech spectrum, and the vibration of vocal folds in the larynx, which function as a pressure-controlled valve regulating airflow into the the vocal tract. This research explores the application of the generalized dynamic pressure-control valve model to the synthesis of voic...
متن کاملVoiced Speech Synthesis Using Pitch Asynchronous Code Excited Linear Filters for the Glottal Source
This paper proposes a model for natural quality voiced speech synthesis using code excited linear all-pole filter for modeling the glottal source signal. Classical glottal signal models are explicit-time functions which inhibit joint sourcetract parameter estimation and require pitch synchronous estimation with precise segmentation of open and closed glottis phase. These problems are overcome i...
متن کاملSource-filter separation for articulation-to-speech synthesis
In this paper we examine a method for separating out the vocal-tract filter response from the voice source characteristic using a large articulatory database. The method realises such separation for voiced speech using an iterative approximation procedure under the assumption that the speech production process is a linear system composed of a voice source and a vocal-tract filter, and that each...
متن کاملInvestigating source and filter contributions, and their interaction, to statistical parametric speech synthesis
This paper presents an investigation of the separate perceptual degradations introduced by the modelling of source and filter features in statistical parametric speech synthesis. This is achieved using stimuli in which various permutations of natural, vocoded and modelled source and filter are combined, optionally with the addition of filter modifications (e.g. global variance or modulation spe...
متن کاملAutomatic voice-source parameterization of natural speech
We present here our work in automatic parameterization of natural speech by means of a pitch synchronous source-filter decomposition algorithm. The derivative glottal source is modelled using the Liljencrants-Fant (LF) model. The model parameters are obtained simultaneously with the coefficients of an all-pole filter representing the vocal tract response by means of a quadratic programming algo...
متن کامل