Unlock instant, AI-driven research and patent intelligence for your innovation.

Identification of unit overlay region in concatenated speech sound synthesis system

A technology of unit overlap and speech synthesis, applied in speech synthesis, speech recognition, speech analysis, etc., can solve impractical problems

Inactive Publication Date: 2004-07-21
PANASONIC CORP
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, such systems rely entirely on computationally expensive processing, making them impractical for many applications

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Identification of unit overlay region in concatenated speech sound synthesis system
  • Identification of unit overlay region in concatenated speech sound synthesis system
  • Identification of unit overlay region in concatenated speech sound synthesis system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] To better understand the techniques employed in the present invention, a basic understanding of adapter synthesis should be gained. figure 1 The cohesive synthesis process is illustrated with an example in which sound units (in this case syllables) from two different words are joined to form a third word. More specifically, sound units from the words "suffice" and "tight" combine to synthesize the new word "fight".

[0022] refer to figure 1 , extract time series data from the words "suffice" and "tight" to determine the sound units 10 and 12, preferably on syllable boundaries. In this case, the sound unit 10 is further divided at 14 in order to separate out the relevant parts required for splicing.

[0023] The sound units are aligned at 16 so that an overlap zone is defined by corresponding portions 18 and 20 . After alignment, the time series data is merged at 22 to synthesize new words.

[0024] The invention relates specifically to the overlap region 16, and pa...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

Speech signal parameters are extracted from time-series data corresponding to different sound units containing the same vowel. The extracted parameters are used to train a statistical model, such as a Hidden Markov-based Model, that has a data structure for separately modeling the nuclear trajectory region of the vowel and its surrounding transition elements. The model is trained as through embedded re-estimation to automatically determine optimally aligned models that identify the nuclear trajectory region. The boundaries of the nuclear trajectory region serve to delimit the overlap region for subsequent sound unit concatenation.

Description

technical field [0001] The present invention relates to a cohesive speech synthesis system. In particular, the present invention relates to a system and method for identifying suitable boundary regions when concatenating speech units. The system employs a phonetic unit database using phonetic unit models. Background technique [0002] Different forms of cohesive speech synthesis systems exist today, and they differ in how cohesive speech units are stored and processed. These forms include time-domain waveform representations, frequency-domain representations (such as formant representations or linear predictive coding LPC representations), or some combination thereof. [0003] Regardless of the phonetic unit form, cohesive synthesis is accomplished by identifying appropriate boundary regions at each unit boundary. These units are smoothly overlapped to synthesize new sound units, including words and phrases. The phonetic units in cohesive synthesis systems are usually di...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F15/18G06N3/04G10L13/06G10L13/08G10L15/14G10L15/16
CPCG10L13/07
Inventor 尼古拉斯·基布雷史蒂夫·皮尔逊
Owner PANASONIC CORP