公开(公告)号：US07062438B2

公开(公告)日：2006-06-13

申请号：US10388107

申请日：2003-03-13

申请人： Kenichiro Kobayashi ,  Nobuhide Yamazaki ,  Makoto Akabane

发明人： Kenichiro Kobayashi ,  Nobuhide Yamazaki ,  Makoto Akabane

IPC分类号： G10L13/08

CPC分类号： G10L13/033

摘要： A sentence or a singing is to be synthesized with a natural speech close to the human voice. To this end, singing metrical data are formed in a tag processing unit 211 in a singing synthesis unit 212 in a speech synthesis apparatus 200 based on singing data and an analyzed text portion. A language analysis unit 213 performs language processing on text portions other than the singing data. As for a text portion registered in a natural metrical dictionary, as determined by this language processing, corresponding natural metrical data is selected and its parameters are adjusted in a metrical data adjustment unit 222 based on phonemic segment data of a phonemic segment storage unit 223 in the metrical data adjustment unit 222. As for a text portion not registered in the natural metrical dictionary, a phonemic symbol string is generated in a natural metrical dictionary storage unit 214, after which metrical data are generated in a metrical generating unit 221. A waveform generating unit 224 concatenates necessary phonemic segment data, based on the natural metrical data, metrical data and the singing metrical data to generate speech waveform data.

摘要翻译： 一句话或一首歌唱是用一种接近人声的自然语言来合成的。 为此，在语音合成装置200的唱歌合成装置200中，基于唱歌数据和分析文本部分，在标签处理部211中形成歌唱数据。 语言分析单元213对唱歌数据以外的文本部分进行语言处理。 对于通过该语言处理确定的自然语言字典中登记的文本部分，根据音素段存储单元223的音素段数据，选择对应的自然计量数据并在数字数据调整单元222中调整其参数 计量数据调整单元222。 对于没有登记在自然计量词典中的文本部分，在自然计量字典存储单元214中生成音符符号串，之后在计量生成单元221中生成计量数据。 波形生成单元224基于自然的测量数据，测量数据和歌唱数据连接必要的音素段数据，以产生语音波形数据。

公开(公告)号：US5208594A

公开(公告)日：1993-05-04

申请号：US874817

申请日：1992-04-28

申请人： Nobuhide Yamazaki

发明人： Nobuhide Yamazaki

IPC分类号： H03M3/02 ,  H03M3/04

CPC分类号： H03M3/50 ,  H03M3/458

摘要： A signal processing system comprises a delta-sigma modulation unit supplied with an input analog signal for producing one-bit data as a result of a delta-sigma modulation, an arithmetic unit supplied with the output one-bit data of the delta-sigma modulation unit for applying a predetermined arithmetic operation thereon, an integration unit for integrating an output of the arithmetic unit to produce multiple-bit data as an output, a comparator unit supplied with the multiple-bit data from the integration unit at a first input port and multiple-bit reference data at a second input port for producing one-bit data as a result of comparison, a feedback unit supplied with the output of the comparator unit for producing the multiple-bit reference data based upon the one-bit data produced by the comparator unit such that said reference data predicts the multiple-bit data supplied to the first input port, and a digital-to-analog conversion unit supplied with said output one-bit data of the comparator unit for converting the same to an analog output signal.

摘要翻译： 信号处理系统包括一个Δ-Σ调制单元，其被提供有用于产生作为Δ-Σ调制的结果的一位数据的输入模拟信号，提供有Δ-Σ调制的输出一位数据的算术单元 用于对其进行预定算术运算的单元，用于对运算单元的输出进行积分以产生多位数据作为输出的积分单元，在第一输入端口从积分单元提供多比特数据的比较单元， 作为比较结果的用于产生1比特数据的第二输入端口的多比特参考数据，提供有比较器单元的输出的反馈单元，用于根据由比特单元产生的1比特数据产生多比特参考数据 所述比较器单元使得所述参考数据预测提供给所述第一输入端口的多位数据，以及提供有所述com的所述输出的一位数据的数模转换单元 用于将其转换为模拟输出信号的解算器单元。

公开(公告)号：US20050010414A1

公开(公告)日：2005-01-13

申请号：US10862656

申请日：2004-06-07

申请人： Nobuhide Yamazaki

发明人： Nobuhide Yamazaki

IPC分类号： G10L13/00 ,  G10L11/04 ,  G10L13/04 ,  G10L13/06 ,  G10L21/04

CPC分类号： G10L13/04 ,  G10L25/24 ,  G10L2013/021

摘要： A speech synthesis apparatus and a speech synthesis method, in which a waveform of a desired formant shape may be generated with a small volume of computing operations. A voiced sound generating unit of the speech synthesis apparatus includes n single formant generating units, an adder for summing these outputs to generate a one-pitch waveform, a one-pitch buffer unit, and a waveform overlapping unit for overlapping a number of the one-pitch waveforms as the one-pitch waveform is shifted by one pitch period each time. Each single formant generating unit is supplied with three parameters, namely a center frequency of a formant representing the formant position, a formant bandwidth, and a formant gain and reads out the band characteristics waveform at a readout interval, derived from the bandwidth wn, from a band characteristics waveform storage unit to effect expansion along the time axis. The resulting waveform is multiplied with a sine wave of the center frequency to output a pitch waveform for a formant representing characteristics of a formant.

摘要翻译： 一种语音合成装置和语音合成方法，其中可以以少量的计算操作产生期望的共振峰形状的波形。 语音合成装置的有声声音产生单元包括n个单一共振峰发生单元，用于将这些输出相加以产生单音调波形的加法器，一个音调缓冲单元和用于重叠一个音调波形的波形重叠单元 一次音调波形每次偏移一个音调周期。 向每个共振峰发生单元提供三个参数，即表示共振峰位置的共振峰的中心频率，共振峰带宽和共振峰增益，并从带宽wn导出的读出间隔读出频带特性波形， 带状特性波形存储单元，沿着时间轴进行扩展。 所得到的波形与中心频率的正弦波相乘以输出表示共振峰特征的共振峰的音调波形。

公开(公告)号：US06178442B1

公开(公告)日：2001-01-23

申请号：US08868723

申请日：1997-06-04

申请人： Nobuhide Yamazaki

发明人： Nobuhide Yamazaki

IPC分类号： G06F1516

CPC分类号： G06Q10/107 ,  H04L51/30

摘要： Identification information, access information indicating that a main body of a mail message has not been accessed nor read at a destination terminal, information about the sender, information about the receiver, time and date information, and subject information each collated to each other is stored and managed as a transmitted mail managing information in a mail box. When a mail message is transmitted, a transmitted mail message including a return mail program actuated to return a response, when the transmitted mail message is accessed and read at the destination terminal, an acknowledgement that the transmitted mail message has been accessed and read and identification information for the transmitted mail message as a returned mail to a server. When information as to whether the transmitted mail message has been accessed and read or not is obtained from the received return mail message, the identification information is extracted from the received return mail message and the access information for the transmitted mail message corresponding to the identification information is changed to information indicating that the transmitted mail message has been accessed and read.

摘要翻译： 识别信息，指示邮件主体的主体尚未被访问的访问信息或在目的地终端处读取，存储关于发送者的信息，关于接收者的信息，时间和日期信息以及彼此对照的对象信息 并作为邮件中的传送邮件管理信息进行管理。 当发送邮件消息时，发送的邮件消息包括被启动以返回响应的返回邮件程序，当在目的地终端上访问和读取所发送的邮件消息时，确认发送的邮件消息已经被访问和识别 将发送的邮件消息作为返回的邮件发送到服务器的信息。 当从接收的返回邮件消息中获得关于发送的邮件消息是否被访问和不被读取的信息时，从接收到的返回邮件消息中提取标识信息，以及与所述标识信息相对应的所发送的邮件消息的访问信息 被更改为指示所发送的邮件消息已被访问和读取的信息。

公开(公告)号：US5875427A

公开(公告)日：1999-02-23

申请号：US828942

申请日：1997-03-28

申请人： Nobuhide Yamazaki

发明人： Nobuhide Yamazaki

IPC分类号： G10L13/04 ,  G10L21/06 ,  G10L5/02

CPC分类号： G10L21/06 ,  G10L13/00

摘要： A voice-generating information making apparatus comprises: a talking way data storing section for storing therein talking way data comprising character string information grouped according to the character string information, a character string input unit for inputting a character string (consisting of a control section, an application storing section, a key entry section, and a display section), a retrieving unit for retrieving a group having the same character string information as the inputted character string, a voice tone data storing section for storing therein a plurality of voice tone data, a voice synthesizing section for synthesizing a voice, a voice selecting unit for selecting a desired voice from the synthesized voice, and a voice-generating document storing section for storing therein talking way data corresponding to the selected voice as a voice-generating document in correlation to the inputted character string.

摘要翻译： 一种语音生成信息制作装置，包括：通话方式数据存储部分，用于存储包括根据字符串信息分组的字符串信息的通话方式数据;字符串输入单元，用于输入字符串（由控制部分组成， 应用存储部分，密钥输入部分和显示部分），用于检索具有与输入的字符串相同的字符串信息的组的检索单元，用于在其中存储多个语音音调数据的语音音调数据存储部分 ，用于合成语音的语音合成部分，用于从合成语音中选择所需语音的语音选择单元，以及语音生成文档存储部分，用于存储与所选择的语音相对应的通话方式数据作为语音生成文档 与输入的字符串的相关性。

公开(公告)号：US07596497B2

公开(公告)日：2009-09-29

申请号：US10862656

申请日：2004-06-07

申请人： Nobuhide Yamazaki

发明人： Nobuhide Yamazaki

IPC分类号： G10L13/00 ,  G10L13/06

CPC分类号： G10L13/04 ,  G10L25/24 ,  G10L2013/021

摘要： A speech synthesis apparatus and a speech synthesis method, in which a waveform of a desired formant shape may be generated with a small volume of computing operations. A voiced sound generating unit of the speech synthesis apparatus includes n single formant generating units, an adder for summing these outputs to generate a one-pitch waveform, a one-pitch buffer unit, and a waveform overlapping unit for overlapping a number of the one-pitch waveforms as the one-pitch waveform is shifted by one pitch period each time. Each single formant generating unit is supplied with three parameters, namely a center frequency of a formant representing the formant position, a formant bandwidth, and a formant gain and reads out the band characteristics waveform at a readout interval, derived from the bandwidth wn, from a band characteristics waveform storage unit to effect expansion along the time axis. The resulting waveform is multiplied with a sine wave of the center frequency to output a pitch waveform for a formant representing characteristics of a formant.

摘要翻译： 一种语音合成装置和语音合成方法，其中可以以少量的计算操作产生期望的共振峰形状的波形。 语音合成装置的有声声音产生单元包括n个单一共振峰发生单元，用于将这些输出相加以产生单音调波形的加法器，一个音调缓冲单元和用于重叠一个音调波形的波形重叠单元 一次音调波形每次偏移一个音调周期。 向每个共振峰发生单元提供三个参数，即表示共振峰位置的共振峰的中心频率，共振峰带宽和共振峰增益，并从带宽wn导出的读出间隔读出频带特性波形， 带状特性波形存储单元，沿着时间轴进行扩展。 所得到的波形与中心频率的正弦波相乘以输出表示共振峰特征的共振峰的音调波形。

公开(公告)号：US6088674A

公开(公告)日：2000-07-11

申请号：US821078

申请日：1997-03-20

申请人： Nobuhide Yamazaki

发明人： Nobuhide Yamazaki

IPC分类号： G10L13/04 ,  G10L7/02

CPC分类号： G10L13/04

摘要： Voice-generating information, comprising discrete voice data for velocity or pitch of a voice is made by dispensing the discrete data so that the voice data is not dependent on a time lag between phonemes and at the same time is present at a relative level against a reference thereof. The said information includes data on plural types of voice tone, and is stored in a voice-generating information storing section. Voice tone data indicating sound parameters for each voice element, such as phoneme for each voice tone type, is stored in a voice tone storing section. Voice data, corresponding to the type of voice tone in the voice-generating information stored in the voice-generating storing section, is selected from a plurality of voice type data stored in the voice tone storing section under control by a control section. Meter patterns, which occur successively in the direction of a time axis, are developed according to the voice-generating information. A voice waveform is synthesized according to the meter patterns and to the selected voice tone data with the voice outputted from a speaker.

摘要翻译： 通过分配离散数据，使语音数据不依赖于音素之间的时滞，并且同时存在于相对于相对电平的相对电平，从而产生用于语速的速度或音调的离散语音数据的声音产生信息 参考。 所述信息包括关于多种类型的语音音调的数据，并且被存储在语音产生信息存储部分中。 指示每个语音元素（例如每个语音类型的音素）的声音参数的语音数据被存储在语音音调存储部分中。 从存储在声音存储部分中的多个语音类型数据中，由控制部分控制，从存储在语音产生存储部分中的语音产生信息中的语音音调类型对应的语音数据。 根据语音产生信息显示在时间轴方向上连续发生的电表模式。 根据仪表模式和从扬声器输出的声音对所选择的语音音数据进行语音波形的合成。

公开(公告)号：US20050119889A1

公开(公告)日：2005-06-02

申请号：US10864130

申请日：2004-06-09

申请人： Nobuhide Yamazaki

发明人： Nobuhide Yamazaki

IPC分类号： G10L13/06 ,  G10L13/08 ,  G10L15/08 ,  G10L15/12

CPC分类号： G10L13/07

摘要： A rule based speech synthesis apparatus by which concatenation distortion may be less than a preset value without dependency on utterance, wherein a parameter correction unit reads out a target parameter for a vowel from a target parameter storage, responsive to the phoneme at the a leading end and at a trailing end of a speech element and acoustic feature parameters output from a speech element selector, and accordingly corrects the acoustic feature parameters of the speech element. The parameter correction unit corrects the parameters, so that the parameters ahead and behind the speech element are equal to the target parameter for the vowel of the corresponding phoneme, and outputs the so corrected parameters.

摘要翻译： 一种基于规则的语音合成装置，其中连接失真可以小于预设值而不依赖于话语，其中参数校正单元响应于前端的音素从目标参数存储器中读出元音的目标参数 并且在语音元素的尾端和从语音元素选择器输出的声学特征参数，并且因此校正语音元素的声学特征参数。 参数校正单元校正参数，使语音元素前后的参数等于相应音素元音的目标参数，并输出如此校正的参数。

公开(公告)号：US5864814A

公开(公告)日：1999-01-26

申请号：US828643

申请日：1997-03-31

申请人： Nobuhide Yamazaki

发明人： Nobuhide Yamazaki

IPC分类号： G10L13/033 ,  G10L13/06 ,  G10L5/04

CPC分类号： G10L13/033 ,  G10L13/06

摘要： An information communication system, having host and remote terminal devices, and method for generating a voice in which one voice tone data is selected from a plurality of types of voice tone data and stored according to received voice generating information. The voice is reproduced by generating a voice waveform according to a meter pattern and selected voice tone data. The discrete voice data may be presented for either one or both of velocity and pitch of a voice correlated to a time lag between discrete voice data. The discrete data is dispensed so that each voice data is not dependent on a time lag between phonemes and at the same time is present at a level relative to a reference value. Voice tone data indicating a sound parameter for each voice element such as a phoneme for each voice tone type is stored in a voice tone data storing section in a terminal device. File information is transferred from a host device to a terminal device according to a request from the terminal device, and the terminal device reads out voice tone data specified by the voice-generating information in the file information thereto from a voice tone storing section. A voice is synthesized according to the voice tone data and the voice generating information.

摘要翻译： 具有主机和远程终端设备的信息通信系统以及用于生成语音的方法，其中从多种类型的语音音调数据中选择一种语音音调数据，并根据接收到的语音产生信息进行存储。 通过根据仪表图案和选择的语音音数据产生语音波形来再现语音。 离散语音数据可以呈现出与离散语音数据之间的时间延迟相关的语音的速度和音高之一或两者。 离散数据被分配，使得每个语音数据不依赖于音素之间的时间延迟，并且同时存在于相对于参考值的水平。 指示用于每个语音音调类型的音素的每个语音元素的声音参数的语音音频数据被存储在终端设备中的语音音调数据存储部分中。 文件信息根据来自终端设备的请求从主机设备传送到终端设备，终端设备从语音存储部分读出由文件信息中的语音产生信息指定的语音音频数据。 根据语音音数据和语音产生信息合成语音。

公开(公告)号：US5121428A

公开(公告)日：1992-06-09

申请号：US610317

申请日：1990-11-08

申请人： Hiroki Uchiyama ,  Hiroo Kitagawa ,  Nobuhide Yamazaki

发明人： Hiroki Uchiyama ,  Hiroo Kitagawa ,  Nobuhide Yamazaki

IPC分类号： G10L17/00

CPC分类号： G10L17/02

摘要： In a speaker verification system, a detecting part detects a speech section of an input speech signal by using a time-series acoustic parameters thereof. A segmentation part calculates individuality information for segmentation by using the time-series acoustic parameters within the speech section, and segments the input speech section into a plurality of blocks based on the individuality information. A feature extracting part extracts features of an unknown speaker for every segmented block by using the time-series acoustic parameters. A distance calculating part calculates a distance between the features of the speaker extracted by the feature extracting part and reference features stored in a memory. A decision part makes a decision as to whether or not the unknown speaker is a real speaker by comparing the calculated distance with a predetermined threshold value. Segmentation is made by calculating a primary moment of the spectrum, over a block, and finding successive values which satisfy a predetermined criterion.

摘要翻译： 在扬声器验证系统中，检测部分通过使用其时间序列声学参数来检测输入语音信号的语音部分。 分割部分通过使用语音部分内的时间序列声学参数来计算用于分割的个体信息，并且基于个性信息将输入语音部分分割成多个块。 特征提取部分通过使用时间序列声学参数来提取每个分段块的未知扬声器的特征。 距离计算部分计算由特征提取部分提取的扬声器的特征与存储在存储器中的参考特征之间的距离。 决定部件通过将计算出的距离与预定阈值进行比较，来决定未知扬声器是否是真实的扬声器。 通过在块上计算光谱的主要时刻并找到满足预定标准的连续值来进行分割。