Journal ArticleDOI
Speech synthesis from text
Reads0
Chats0
TLDR
Text analysis for speech synthesis is described in relation to the information needed in speech production, including a pronouncing dictionary and letter-to-sound rules, morphological analysis and accent assignment, and syntactic analysis.Abstract:
Text analysis for speech synthesis is described in relation to the information needed in speech production. This includes a pronouncing dictionary and letter-to-sound rules, morphological analysis and accent assignment, and syntactic analysis. Prosody control rules (fundamental frequency control and segmental duration control) are examined. Speech units for synthesis and parametric representation of speech signals are discussed. Applications and development tools are considered. >read more
Citations
More filters
Journal ArticleDOI
A rose is a REEZ: The two-cycles model of phonology assembly in reading English
Iris Berent,Charles A. Perfetti +1 more
TL;DR: This article proposed a model of phonological assembly that postulates a multilinear representation that segregates consonants and vowels in different planes, which determines the online process of assembly.
Patent
Electronic news reception apparatus that selectively retains sections and searches by keyword or index for text to speech conversion
TL;DR: In this article, an electronic news receiving device receives text data for an electronic edition of a newspaper in the evening and audibly reads the newspaper to the user the next day, while the retained news articles are stored in memory.
PatentDOI
Automated voice synthesis employing enhanced prosodic treatment of text, spelling of text and rate of annunciation
TL;DR: In this paper, prosodic shaping of text sequences appropriate for the discourse in large groupings of text segments, with prosodic boundaries developed to indicate conceptual units within the text groupings, is implemented in a preferred embodiment.
PatentDOI
Methods for controlling the generation of speech from text representing names and addresses
TL;DR: Performance enhancement of the underlying text comprehensibility is obtained through prosodic treatment of the synthesized material, improved speaking rate treatment, and improved methods of spelling words or terms for the system user.
Journal ArticleDOI
Interacting with computers by voice: automatic speech recognition and synthesis
TL;DR: This paper examines how people communicate with computers using speech, and the popular mathematical model called the hidden Markov model (HMM) is examined; first-order HMMs are efficient but ignore long-range correlations in actual speech.
References
More filters
Journal ArticleDOI
Review of text‐to‐speech conversion for English
TL;DR: This review traces the early work on the development of speech synthesizers, discovery of minimal acoustic cues for phonetic contrasts, evolution of phonemic rule programs, incorporation of prosodic rules, and formulation of techniques for text analysis.
Journal ArticleDOI
Declination ‘‘reset’’ and the hierarchical organization of utterances
TL;DR: The authors found that the differences in boundary strength (but‐boundary stronger than and ) would be reflected in the way declination was reset following the boundaries, and suggested the incorporation of hierarchical information into models that analyze F0 contours as strings of abstract targets, which makes problems for models of F0 in which contours result from the interaction of a number of preplanned overall trends.
Book
The organization of Japanese prosody
Abstract: This paper discusses several topics concerning what I call the ,, semantic constraint" on Japanese compounds, a constraint which blocks the prosodic compound formation process in the language. The main discussion starts with the claim that there are two types of compounds in Japanese, "compounding compounds, " to which the prosodic compound rule readily applies, and "non-compounding compounds", which somehow fail to undergo the process. After justifying this position, I will attempt to make a detailed examination of the marked semantic structures constituting the "semantic constraint, " which are responsible for the second type of compounds. I will also show the fact that the compound formation process in English admits of exceptions which are significantly similar to those of Japanese. The second part of the paper focuses on the nature and role of the "semantic constraint" in "complex" compound nouns, or compound nouns consisting of three or more elements. It will be shown that this constraint enables us to uncover the regularities which these compounds exhibit in accentual patterning. -i pý. 'fiý#, , 26 -iZ. MZ
Proceedings ArticleDOI
A diphone synthesis system based on time-domain prosodic modifications of speech
TL;DR: A novel time-domain algorithm is presented for text-to-speech synthesis using diphone concatenation based on the pitch-synchronous overlap-add (PSOLA) approach and is capable of good quality prosodic modifications of natural speech.
Proceedings ArticleDOI
Speech synthesis by rule using an optimal selection of non-uniform synthesis units
TL;DR: This investigation provides an estimate of an appropriate number of Japanese phoneme sequences in this synthesis scheme, which has two advantages in the usage of speech segment units: flexible use of nonuniform synthesis units and the optimal choice of a unit sequence for an input phoneme string using appropriateness measures.