Prosody control in HMM-based speech synthesis
نویسنده
چکیده
In HMM-based speech synthesis, trained statistical models (context-dependent HMMs) are used to predict duration and generate parameters like mel-cepstral coefficients, log F0 values, and bandpass voicing strengths using the maximum likelihood parameter generation algorithm including global variance (Toda et al, 2007). In the later stages, F0 parameters, bandpass voicing strengths, and the five bandpass filters are used to generate a mixed excitation signal. Finally, speech is synthesized from the mel-cepstral coefficients and the mixed excitation signal using the MLSA filter.
منابع مشابه
Evaluation of Finnish unit selection and HMM-based speech synthesis
Unit selection and hidden Markov model (HMM) based synthesis have become the dominant techniques in text-to-speech (TTS) research. In this work, we combine HMM-based signal generation with the front end originally designed for unit selection based Finnish TTS and we evaluate the prosody of the output generated by the two synthesis techniques using the same speech database. Furthermore, we study...
متن کاملPerformance Analysis of Text To Speech Synthesis System Using HMM And Prosody Features With Parsing For Tamil Language
This paper describes a Hidden Markov Model (HMM) based (TTS) system and prosody based (TTS) system for producing natural sounding synthetic speech in Tamil language. The (HMM) based system consists of two phases such as training and synthesis. Tamil speech is first parameterized into spectral and excitation features using Glottal Inverse Filtering (GIF). An emotions present in the input text is...
متن کاملSuperpositional Modeling of Fundamental Frequency Contours for HMM-based Speech Synthesis
Statistical parametric speech synthesis technologies, such as HMM-based and DNN-based ones, gain special attention from researchers because of their ability in generating speech in various voice qualities and styles. In these methods, all acoustic parameters (except durational ones) are handled in a frame-by-frame manner, which is not appropriate for prosodic features. Although relation of adja...
متن کاملUsing FO Contour Generation Process Model for Improved and Flexible Control of Prosodie Features in HMM-based Speech Synthesis
Generation process model of fundamental frequency contours known as Fujisaki's model is ideal to represent global features of prosody. It is a command response model, where the commands have clear relations with linguistic and para/non linguistic information included in the utterance. Therefore, by controlling fundamental frequency contours in the framework of the generation process model, a mo...
متن کاملSyllable based models for prosody modeling in HMM based speech synthesis
Simple4All is a speech synthesis research project that aims to ease the production of synthetic voices in new languages by means of unsupervised modeling techniques. In this work, we introduce syllable based models for prosody modeling in Hidden Markov Model based Text-to-Speech system (HTS). As a part of investigating the potential for building speech synthesis systems in new languages with li...
متن کامل