专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明专利

JPWO2015125329A1 マルチモーダル情報処理装置有权
公开(公告)号：JPWO2015125329A1
公开(公告)日：2017-03-30
申请号：JP2016503921
申请日：2014-09-18
申请人：三菱電機株式会社
发明人：啓吾川島 , 啓吾川島 , 貴弘大塚 , 貴弘大塚 , 山浦　正 , 正山浦
IPC分类号： G06F3/0484
CPC分类号： G10L15/1815 , G06F3/038 , G06F2203/0381 , G10L2015/223 , H04L12/2803 , H04N21/43615
摘要：出力情報生成部3は、入力部1からの情報に対応して出力部5a,5b,…の種類とは無関係の抽象化出力情報4を生成する。出力部5a,5b,…の意味解釈部54a,54b,…は、自出力部の動作状況を監視する状況監視部51a,51b,…の監視結果に基づいて、抽象化出力情報4から具現化出力情報55a,55b,…を生成する。処理実行部56a,56b,…は、具現化出力情報55a,55b,…に対応した処理を行う。

2. 发明专利

JP5322793B2 How speech synthesis apparatus and the speech synthesis 有权
公开(公告)号：JP5322793B2
公开(公告)日：2013-10-23
申请号：JP2009143326
申请日：2009-06-16
申请人：三菱電機株式会社
发明人：正山浦 , 裕久田崎
IPC分类号： G10L13/06

3. 发明专利

JP5089473B2 How speech synthesis apparatus and the speech synthesis 有权
公开(公告)号：JP5089473B2
公开(公告)日：2012-12-05
申请号：JP2008109190
申请日：2008-04-18
申请人：三菱電機株式会社
发明人：正山浦 , 訓古田 , 貴弘大塚 , 裕久田崎
IPC分类号： G10L13/06

4. 发明专利

JP4805121B2 Voice synthesizer, speech synthesis method and a speech synthesis program 有权
公开(公告)号：JP4805121B2
公开(公告)日：2011-11-02
申请号：JP2006340235
申请日：2006-12-18
申请人：三菱電機株式会社
发明人：訓古田 , 貴弘大塚 , 正山浦
IPC分类号： G10L13/08
摘要： PROBLEM TO BE SOLVED: To prevent voice quality degradation by preventing missing of side-lobe waveform information, even when a pitch frequency of synthesized voice is made high. SOLUTION: A window function P W is set, which has a window length corresponding to a pitch cycle P pro indicated by rhythm information which is output from a rhythm setting section 4. By using the window function, a waveform clip section 7 for clipping the pitch waveform from the waveform of a phoneme which is selected by a phoneme selection section 6 is provided, and a voice synthesizing section 8 creates synthesized voice of a text, according to a pitch waveform clipped by the pitch cycle P pro indicated by the rhythm information which is output from the rhythm setting section 4, and the pitch waveform clipped by the waveform clip section 7. COPYRIGHT: (C)2008,JPO&INPIT

5. 发明专利

JP4393794B2 Voice synthesizer 有权
公开(公告)号：JP4393794B2
公开(公告)日：2010-01-06
申请号：JP2003154989
申请日：2003-05-30
申请人：三菱電機株式会社
发明人：正山浦 , 真哉高橋
IPC分类号： G10L13/06 , G10L19/00

6. 发明专利

JP5595605B2 Audio signal restoring apparatus and an audio signal restoration method 有权
公开(公告)号：JP5595605B2
公开(公告)日：2014-09-24
申请号：JP2013551031
申请日：2011-12-27
申请人：三菱電機株式会社
发明人：耕佑細谷 , 訓古田 , 正山浦
IPC分类号： G10L21/0388 , G10L21/0232
CPC分类号： G10L19/012 , G10L21/0208 , G10L21/0388

7. 发明专利

JP4907677B2 Speech coding apparatus and speech encoding method 有权
公开(公告)号：JP4907677B2
公开(公告)日：2012-04-04
申请号：JP2009018494
申请日：2009-01-29
申请人：三菱電機株式会社
发明人：正山浦 , 裕久田崎
IPC分类号： G10L19/08 , G10L19/12
摘要： PROBLEM TO BE SOLVED: To prevent quality deterioration when a pitch period is different from a repetition period. SOLUTION: A period preselecting means 23 finds repetition period candidates of a plurality of driving sound sources by multiplying the repetition period of an adaptive sound source by a plurality of constants and selects repetition period candidates of a predetermined number of driving sound sources. A driving sound source encoding means 27 outputs the evaluation values of sound source positions where encoding distortion is minimized, polarities and the encoding distortions at that time by the repetition period candidates of the predetermined number of driving sound sources. A period encoding means 28 compares the evaluation values of the encoding distortions by the repetition periods, selects a repetition period candidate of the driving sound source based on the comparison result, and outputs selection information, a sound position code and a polarity. COPYRIGHT: (C)2009,JPO&INPIT

8. 发明专利

JP4869882B2 Speech decoding apparatus 有权
公开(公告)号：JP4869882B2
公开(公告)日：2012-02-08
申请号：JP2006316871
申请日：2006-11-24
申请人：三菱電機株式会社
发明人：渉伏見 , 勇小川 , 正山浦 , 久矢島 , 茂明鈴木
IPC分类号： G10L19/00
摘要： PROBLEM TO BE SOLVED: To solve the problem that although a speech is broken during speech communication when an underflow occurs to the buffer of a PC sound card for PC operation of a speech decoding device, data interpolation wherein the underflow of the buffer of the PC sound card is taken into consideration is not performed and a decoding processing load increases when a fluctuation absorbing buffer control system is applied to the buffer control system of the PC sound card. SOLUTION: A packet decomposing means decomposes received packets to extract encoded speech data, and stores them in a fluctuation absorbing buffer, and a speech decoding means decodes the encoded speech data in the fluctuation absorbing buffer and outputs a speech signal, which is stored in a speech signal storage buffer. An interpolation data generating means generates interpolation data following the decoded speech signal and adds it to the speech signal storage buffer each time the decoded speech signal is stored in the speech signal storage buffer. COPYRIGHT: (C)2008,JPO&INPIT

9. 发明专利

JP4510977B2 Speech coding method and speech decoding method and apparatus 有权
公开(公告)号：JP4510977B2
公开(公告)日：2010-07-28
申请号：JP2000033843
申请日：2000-02-10
申请人：三菱電機株式会社
发明人：正山浦 , 裕久田崎
IPC分类号： G10L19/08 , G10L19/12 , H03M7/30
摘要： PROBLEM TO BE SOLVED: To improve valuational precision of a noisy degree of a voice and to reproduce a high quality voice. SOLUTION: This device is provided with a noisy degree valuation means 25 for valuating a noisy degree by using the code or an encode result of a gain for an adaptive code vector outputted from an adaptive code book 9 and for a drive code vector outputted from a first drive code book 11 and a second drive code book 12 as one of elements and a noisy degree valuation means 26 for valuating the noisy degree by using the code or a decode result of the gain for an adaptive code vector outputted from an adaptive code book 19 and for a drive code vector outputted from the first drive code book 21 and the second drive code book 22 as one of the elements.

10. 发明专利

JP4087429B2 Speech coding apparatus and speech encoding method 有权
公开(公告)号：JP4087429B2
公开(公告)日：2008-05-21
申请号：JP2007044043
申请日：2007-02-23
申请人：三菱電機株式会社
发明人：正山浦 , 裕久田崎
IPC分类号： G10L19/08 , G10L19/12
摘要： PROBLEM TO BE SOLVED: To prevent quality deterioration when a pitch period is different from a repetition period. SOLUTION: A period preselecting means 23 finds repetition period candidates of a plurality of driving sound sources by multiplying the repetition period of an adaptive sound source by a plurality of constants and selects repetition period candidates of a predetermined number of driving sound sources. A driving sound source encoding means 27 outputs the evaluation values of sound source positions where encoding distortion is minimized, polarities and the encoding distortions at that time by the repetition period candidates of the predetermined number of driving sound sources. A period encoding means 28 compares the evaluation values of the encoding distortions by the repetition periods, selects a repetition period candidate of the driving sound source on the basis of the comparison result, and outputs selection information, a sound position code, and a polarity. COPYRIGHT: (C)2007,JPO&INPIT

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式