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

Speech processing device and method

A voice processing and equipment technology, applied in voice analysis, voice synthesis, voice recognition, etc., can solve problems such as voice processing delay

Active Publication Date: 2018-06-12
FUJITSU LTD
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Synthetic speech generation techniques involve performing high-load processing such as speech recognition using acoustic models, generating phoneme markers, and generating synthetic speech, which can cause delays in speech processing

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
  • Speech processing device and method
  • Speech processing device and method
  • Speech processing device and method

Examples

Experimental program
Comparison scheme
Effect test

no. 1 approach

[0026] figure 1 It is a functional block diagram of the speech processing apparatus 1 according to the first embodiment. The speech processing device 1 includes an acquisition unit 2, a detection unit 3, an accent segment estimation unit 4, a vowel segment length calculation unit 5 (in other words, a vowel segment length specification unit 5), and a control unit 6.

[0027] For example, the acquisition unit 2 is a hardware circuit including wiring logic. The acquiring unit 2 may also be a functional module implemented by a computer program executed in the voice processing device 1. For example, the acquisition unit 2 acquires the input voice via a wired circuit or a wireless circuit. Optionally, for example, the acquiring unit 2 may acquire the input voice from a microphone not shown, which is connected to the voice processing device 1 or is located in the voice processing device 1. For example, the input voice is English, but the input voice can also be any other language. Si...

no. 2 approach

[0084] The first embodiment describes a speech processing device and a speech processing device in which the control unit 6 controls the length of the first vowel segment or the length of the second vowel segment based on the ratio or difference between the length of the first vowel segment and the length of the second vowel segment. Processing method and voice processing program. The second embodiment will describe a voice processing device, a voice processing method, and a voice processing program in which the first vowel segment length and the second vowel segment length are controlled according to the vowel segment length. Since the functional blocks of the voice processing device 1 in the second embodiment are figure 1 The same as in the first embodiment, so only the difference from the first embodiment will be described.

[0085] The control unit 6 implements control to extend the length of the first vowel segment or shorten the length of the second vowel segment. Picture ...

no. 3 approach

[0088] Picture 11 It is a functional block diagram of the speech processing apparatus 1 according to the third embodiment. The speech processing device 1 includes an acquisition unit 2, a detection unit 3, an accent segment estimation unit 4, a vowel segment length calculation unit 5, a control unit 6 and a feature calculation unit 7. Since the acquisition unit 2, the detection unit 3, the accent segment estimation unit 4, the vowel segment length calculation unit 5, and the control unit 6 have functions similar to those in the first embodiment, their detailed description is omitted.

[0089] For example, the feature calculation unit 7 is a hardware circuit including wiring logic. The feature calculation unit 7 may also be a functional module realized by a computer program executed in the speech processing device 1. The feature calculation unit 7 receives the input voice from the acquisition unit 2 and receives the first vowel segment length and the second vowel segment length ...

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

Provided are a voice processing device and a voice processing method. The speech processing device includes a computer processor, and the device includes: an acquisition unit configured to obtain an input speech; a detection unit configured to detect vowel segments contained in the input speech; an accent segment estimation unit configured to Configured to estimate an accent segment contained in the input speech; a vowel segment length specifying unit configured to specify a first vowel segment length containing the accent segment and a second vowel segment length not containing the accent segment; and a control unit , configured to control at least one of the first vowel segment length and the second vowel segment length.

Description

Technical field [0001] The embodiments discussed herein relate to, for example, a voice processing device, a voice processing method, and a voice processing program for controlling input signals. Background technique [0002] For example, with the latest development and internationalization of information processing equipment, it has become more and more common to make telephone calls in foreign languages ​​through telephone applications installed in personal computers. In view of this trend, a method for controlling a voice signal from a non-native speaker of a certain language so that his / her voice can be more easily understood by the native speaker of the language is disclosed. For example, Japanese Patent No. 4942860 discloses a technique for generating a phoneme mark corresponding to an input voice through speech recognition using an acoustic model, converting the phoneme mark according to a specific conversion table, and according to the converted phoneme mark Produce synt...

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): G10L13/08G10L25/87
CPCG10L21/057G10L15/02G10L15/04G10L21/02G10L13/027G10L15/08G10L21/0364
Inventor 外川太郎盐田千里大谷猛
Owner FUJITSU LTD