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. 发明申请

WO2022240449A1 END-TO-END SPEECH DIARIZATION VIA ITERATIVE SPEAKER EMBEDDING 审中-公开
公开(公告)号：WO2022240449A1
公开(公告)日：2022-11-17
申请号：PCT/US2021/070748
申请日：2021-06-22
申请人： GOOGLE LLC
发明人： GRANGIER, David , ZEGHIDOUR, Neil , TEBOUL, Oliver
IPC分类号： G10L25/87 , G10L25/78 , G10L17/18 , G06N3/04 , G10L15/063 , G10L15/07 , G10L19/008
摘要： A method (600) includes receiving an audio signal (122) corresponding to utterances (120) spoken by multiple speakers (10). The method also includes encoding the input audio signal into a sequence of T temporal embeddings (220). During each of a plurality of iterations each corresponding to a respective speaker of the multiple speakers, the method includes selecting a respective speaker embedding (240) for the respective speaker by determining a probability that the corresponding temporal embedding includes a presence of voice activity by a single new speaker for which a speaker embedding was not previously selected during a previous iteration and selecting the respective speaker embedding for the respective speaker as the temporal embedding. The method also includes, at each time step, predicting a respective voice activity indicator (262) for each respective speaker of the multiple speakers based on the speaker embeddings selected and the temporal embedding.

2. 发明申请

WO2021119014A1 SPEECH RECOGNITION THROUGH DISAMBIGUATION FEEDBACK 审中-公开
公开(公告)号：WO2021119014A1
公开(公告)日：2021-06-17
申请号：PCT/US2020/063811
申请日：2020-12-08
申请人： AMAZON TECHNOLOGIES, INC.
发明人： REINSPACH, Julia , ROKHLENKO, Oleg , GOTTUMUKKALA, Ramakanthachary , CLEMENTE, Giovanni , AGRAWAL, Ankit , BHARDWAJ, Swayam , MICHAELI, Guy , KRISHNAMOORTHY, Vaidyanathan Puthucode , VLACHOS, Costantino , VINODKRISHNAN, Nalledath P. , VICKERS, Shaun M. , RAMACHANDRAN, Sethuraman , MOORE, Charles C.
IPC分类号： G10L15/22 , G10L15/06 , G10L15/07 , G06F3/16
摘要： A request including audio data is received from a voice-enabled device. A string of phonemes present in the utterance is determined through speech recognition. At a later time, a subsequent user input corresponding to the request may be received, in which the user input is associated with one or more text keywords. The subsequent user input may be obtained in response to an active request. Alternatively, feedback may not be actively elicited, but rather collected passively. However it is obtained, the one or more keywords associated with the subsequent user input may be associated with the string of phonemes to indicate that the user is saying or mean those words when they product that string of phonemes. A user-specific speech recognition key for the user account is then updated to associate the string of phonemes with these words. A general speech recognition model can also be trained using the association.

3. 发明申请

WO2020214298A1 ADVERSARIAL SPEAKER ADAPTATION 审中-公开
公开(公告)号：WO2020214298A1
公开(公告)日：2020-10-22
申请号：PCT/US2020/022875
申请日：2020-03-16
申请人： MICROSOFT TECHNOLOGY LICENSING, LLC
发明人： MENG, Zhong , LI, Jinyu , GONG, Yifan
IPC分类号： G10L15/07 , G10L15/16 , G10L25/30
摘要： Embodiments are associated with a speaker-independent acoustic model capable of classifying senones based on input speech frames and on first parameters of the speaker-independent acoustic model, a speaker-dependent acoustic model capable of classifying senones based on input speech frames and on second parameters of the speaker-dependent acoustic model, and a discriminator capable of receiving data from the speaker-dependent acoustic model and data from the speaker-independent acoustic model and outputting a prediction of whether received data was generated by the speaker-dependent acoustic model based on third parameters. The second parameters are initialized based on the first parameters, the second parameters are trained based on input frames of a target speaker to minimize a senone classification loss associated with the second parameters, a portion of the second parameters are trained based on the input frames of the target speaker to maximize a discrimination loss associated with the discriminator, and the third parameters are trained based on the input frames of the target speaker to minimize the discrimination loss.

4. 发明申请

WO2020214259A1 CONDITIONAL TEACHER-STUDENT LEARNING FOR MODEL TRAINING 审中-公开
公开(公告)号：WO2020214259A1
公开(公告)日：2020-10-22
申请号：PCT/US2020/020716
申请日：2020-03-03
申请人： MICROSOFT TECHNOLOGY LICENSING, LLC
发明人： MENG, Zhong , LI, Jinyu , ZHAO, Yong , GONG, Yifan
IPC分类号： G10L15/06 , G10L15/065 , G10L15/07 , G10L15/16 , G06N3/08
摘要： Embodiments are associated with conditional teacher-student model training. A trained teacher model configured to perform a task may be accessed and an untrained student model may be created. A model training platform may provide training data labeled with ground truths to the teacher model to produce teacher posteriors representing the training data. When it is determined that a teacher posterior matches the associated ground truth label, the platform may conditionally use the teacher posterior to train the student model. When it is determined that a teacher posterior does not match the associated ground truth label, the platform may conditionally use the ground truth label to train the student model. The models might be associated with, for example, automatic speech recognition (e.g., in connection with domain adaptation and/or speaker adaptation).

5. 发明申请

WO2017166651A1 语音识别模型训练方法、说话人类型识别方法及装置审中-公开
公开(公告)号：WO2017166651A1
公开(公告)日：2017-10-05
申请号：PCT/CN2016/096986
申请日：2016-08-26
申请人：乐视控股(北京)有限公司 , 乐视致新电子科技(天津)有限公司
发明人：张俊博
IPC分类号： G10L15/02 , G10L15/07
CPC分类号： G10L15/02 , G10L15/07
摘要：一种语音识别模型训练方法、说话人类型识别方法及装置，获取训练语音并提取所述训练语音的声学特征（101）；利用所述声学特征，训练获得用于提取说话人特征的特征识别器（102）；利用所述特征识别器从每一个用户类型对应的目标语音中提取说话人特征，作为所述用户类型对应的说话人特征（103）；将不同用户类型对应的说话人特征以及所述特征识别器，作为说话人类型识别模型（104）。利用说话人类型识别模型中所述特征识别器结合待识别语音的声音特征，提取所述待识别语音的说话人特征，并将所述待识别语音的说话人特征与不同用户类型对应的说话人特征进行匹配，将匹配度最高的说话人特征对应的用户类型识别为所述待识别语音的用户类型。

6. 发明申请

WO2017099936A1 SYSTEM AND METHODS FOR ADAPTING NEURAL NETWORK ACOUSTIC MODELS 审中-公开
标题翻译：适应神经网络声学模型的系统和方法
公开(公告)号：WO2017099936A1
公开(公告)日：2017-06-15
申请号：PCT/US2016/061326
申请日：2016-11-10
申请人： NUANCE COMMUNICATIONS, INC.
发明人： ZHAN, Puming , LI, Xinwei
IPC分类号： G10L15/16 , G10L15/07
CPC分类号： G10L15/075 , G10L15/07 , G10L15/14 , G10L15/16 , G10L17/02
摘要： Techniques for adapting a trained neural network acoustic model, comprising using at least one computer hardware processor to perform: generating initial speaker information values for a speaker; generating first speech content values from first speech data corresponding to a first utterance spoken by the speaker; processing the first speech content values and the initial speaker information values using the trained neural network acoustic model; recognizing, using automatic speech recognition, the first utterance based, at least in part on results of the processing; generating updated speaker information values using the first speech data and at least one of the initial speaker information values and/or information used to generate the initial speaker information values; and recognizing, based at least in part on the updated speaker information values, a second utterance spoken by the speaker.
摘要翻译：包括使用至少一个计算机硬件处理器来执行以下操作：调节训练的神经网络声学模型的技术：生成说话者的初始说话者信息值; 从对应于讲话者讲的第一话语的第一讲话数据中产生第一讲话内容值; 使用训练的神经网络声学模型处理第一语音内容值和初始说话者信息值; 至少部分地基于所述处理的结果，使用自动语音识别来识别所述第一话语; 使用第一语音数据和初始说话人信息值和/或用于生成初始说话人信息值的信息中的至少一个生成更新的说话人信息值; 以及至少部分基于更新后的说话者信息值来识别说话者说出的第二话语。

7. 发明申请

WO2016139670A1 SYSTEM AND METHOD FOR GENERATING ACCURATE SPEECH TRANSCRIPTION FROM NATURAL SPEECH AUDIO SIGNALS 审中-公开
标题翻译：用于从自然语音音频信号生成精确语音转录的系统和方法
公开(公告)号：WO2016139670A1
公开(公告)日：2016-09-09
申请号：PCT/IL2016/050246
申请日：2016-03-03
申请人： VOCASEE TECHNOLOGIES LTD
发明人： NIR, Igal
IPC分类号： G10L15/00 , G10L15/04 , G10L15/06 , G10L15/07 , G10L15/02 , G10L15/26 , G10L15/32
CPC分类号： G10L15/08 , G06F17/18 , G06F17/3074 , G10L15/02 , G10L15/05 , G10L15/063 , G10L15/07 , G10L15/32 , G10L25/18
摘要： Apparatus for generating accurate speech transcription from natural speech, comprising a data storage for storing a plurality of audio data items, each of which being recitation of text by a specific speaker! a plurality of ASR modules, each of which being trained to optimally create a unique acoustic/linguistic model according to the spectral components contained in said audio data item and analyzing each audio data item and representing said audio data item by an ASR module! a memory for storing all unique acoustic/linguistic models! a controller, adapted to receive natural speech audio signals and divide each natural speech audio signal to equal segments of a predetermined time! adjust the length of each segment, such that each segment will contain one or more complete words! distribute said segments to all ASR module and activate each ASR module to generate a transcription of the words in each segment according to the level of matching to its unique acoustic/linguistic model! calculate, for each given word in a segment, a confidence measure being the probability that said given word is correct; for each segment and for each ASR module, calculate the average confidence of the transcription; obtain the confidence for each word in the segment and calculating mean confidence value of said word! for each segment, decide which transcription is the most accurate by choose only the ASR module with the highest average confidence, from all chosen ASR modules for said segment and creating the transcription of said audio signal by combining all transcriptions resulting from the decisions made for each segment.
摘要翻译：用于从自然语音产生准确语音转录的装置，包括用于存储多个音频数据项的数据存储器，每个音频数据项由特定扬声器引用文本。多个ASR模块，其中的每一个被训练成根据包含在所述音频数据项中的频谱分量最佳地创建独特的声学/语言模型，并通过ASR模块分析每个音频数据项并表示所述音频数据项！用于存储所有独特的声学/语言模型的记忆！控制器，适于接收自然语音音频信号，并将每个自然语音音频信号划分成预定时间的相等分段！调整每段的长度，使每个段都包含一个或多个完整的单词！将所述段分发到所有ASR模块，并激活每个ASR模块，以根据与其独特的声学/语言模型的匹配水平来生成每个段中的单词的转录！对于段中的每个给定单词，计算所述给定单词正确的概率的置信度度量; 对于每个段和每个ASR模块，计算转录的平均置信度; 获取段中每个单词的置信度，并计算该单词的平均置信度值！对于每个段，通过仅选择具有最高平均置信度的ASR模块，从所述段的所有选择的ASR模块中确定哪个转录是最准确的，并通过组合由每个决定产生的所有转录来创建所述音频信号的转录分割。

8. 发明申请

WO2013185070A1 APPARATUS AND METHODS USING A PATTERN MATCHING SPEECH RECOGNITION ENGINE TO TRAIN A NATURAL LANGUAGE SPEECH RECOGNITION ENGINE 审中-公开
标题翻译：使用模式匹配语音识别引擎训练自然语言语音识别引擎的装置和方法
公开(公告)号：WO2013185070A1
公开(公告)日：2013-12-12
申请号：PCT/US2013/044772
申请日：2013-06-07
申请人： NVOQ INCORPORATED
发明人： CORFIELD, Charles , MARQUETTE, Brian
IPC分类号： G10L15/07 , G10L15/26
CPC分类号： G10L15/07 , G10L15/32
摘要： The technology of the present application provides a speech recognition system with at least two different speech recognition engines or a single engine speech recognition engine with at least two different modes of operation. The first speech recognition being used to match audio to text, which text may be words or phrases. The matched audio and text is used by a training module to train a user profile for a natural language speech recognition engine, which is at least one of the two different speech recognition engines or modes. An evaluation module evaluates when the user profile is sufficiently trained to convert the speech recognition engine from the first speech recognition engine or mode to the natural language speech recognition or mode.
摘要翻译：本申请的技术提供了具有至少两种不同语音识别引擎或具有至少两种不同操作模式的单个引擎语音识别引擎的语音识别系统。第一个语音识别用于将音频与文本进行匹配，哪个文本可以是单词或短语。匹配的音频和文本由训练模块用于训练用于自然语言语音识别引擎的用户简档，所述自然语言语音识别引擎是两种不同语音识别引擎或模式中的至少一种。评估模块评估何时用户简档被充分训练以将语音识别引擎从第一语音识别引擎或模式转换为自然语言语音识别或模式。

9. 发明申请

WO2013180600A2 СПОСОБ ПЕРЕОЗВУЧИВАНИЯ АУДИОМАТЕРИАЛОВ И УСТРОЙСТВО ДЛЯ ЕГО ОСУЩЕСТВЛЕНИЯ 审中-公开
标题翻译：用于重写音频材料的方法和用于执行音频材料的设备
公开(公告)号：WO2013180600A2
公开(公告)日：2013-12-05
申请号：PCT/RU2013/000404
申请日：2013-05-16
申请人： БРЕДИХИН, Александр Юрьевич
发明人： БРЕДИХИН, Александр Юрьевич
IPC分类号： G10L13/02 , G10L15/07
CPC分类号： G10L13/02 , G10L13/033 , G10L21/003
摘要： Способ и устройство позволяют повысить качество проведения фазы обучения, улучшить степень совпадения голоса пользователя (целевого диктора) в конвертированном речевом сигнале, обеспечить возможность одноразового проведения фазы обучения для различных аудиоматериалов. Указанный технический результат достигается тем, что в программно управляемом электронном устройстве обработки информации (ПУЭУОИ) формируют акустическую базу исходных аудиоматериалов (АБИА) и акустическую обучающую базу (АОБ). Передают данные из АБИА для отображения списка исходных аудиоматериалов на экране монитора. При выборе из списка АБИА по меньшей мере одного аудиоматериала, данные о нем передают для сохранения в ОЗУ ПУЭУОИ. Осуществляют выбор из АОБ обучающих фраз диктора файлы, которые преобразуют в звуковые фразы и передают их пользователю на устройство воспроизведения звука. Пользователь посредством микрофона воспроизводит звуковые фразы, в процессе воспроизведения которых на экране монитора отображают текст воспроизводимой фразы и курсор, перемещающийся по тексту фразы в соответствии с тем, как пользователь должен ее воспроизвести. Создают файлы в соответствии с воспроизводимыми фразами, которые сохраняют по порядку воспроизведения фраз в формируемой акустической базе целевого диктора (АБЦД). ПУЭУОИ производит контроль скорости воспроизводимой фразы и ее громкости. Формируют файл функции конверсии. Файлы АБИА, используя файл функции конверсии, конвертируют для сохранения в формируемой акустической базе конвертированных аудиоматериалов (АБКА) и предоставления пользователю данных о конвертированных аудиоматериалах на экране монитора. Устройство содержит соответствующие функциональные блоки, реализующие способ.
摘要翻译：该方法和装置允许增加训练阶段的质量，增加用户（目标讲话者）在转换的语音信号中的语音的一致程度，并提供单次进行训练阶段的可能性各种音响材料。该技术结果通过在软件控制的电子设备中创建原始音频材料的声学数据库和声学训练数据库来实现，用于处理信息。来自原始音频材料的声学数据库的数据用于在监视器的屏幕上显示原始音频材料的列表。在从原始音频材料的声学数据库提供的列表中选择至少一个音频材料之后，将关于所述音频材料的数据转移到存储到软件控制的电子设备的随机存取存储器中以用于处理信息。训练短语文件从声学教学数据库中选出，转化为声学短语，并通过声音再现装置发送给用户。用户通过麦克风再现声学短语，在再现其中的短语的过程中，要再现的短语的文本与沿着短语的文本移动的光标一起显示在监视器屏幕上，根据用户必须重现所述短语的方式。根据要再现的短语创建文件，并按照再现所述文件的顺序将其保存到目标扬声器声学数据库。用于处理信息的软件控制电子设备监视要再现的短语的速度和音量。形成转换功能文件。使用转换功能文件转换原始音频材料的声学数据库的文件，然后将其保存到转换音频材料的声学数据库中，并以转换的音频材料数据的形式呈现给用户显示器的屏幕。该装置包含用于执行该方法的相应功能单元。

10. 发明申请

WO02049006A1 MOBILE TERMINAL CONTROLLABLE BY SPOKEN UTTERANCES 审中-公开
标题翻译：移动终端可通过SPOKEN UTTERANC控制
公开(公告)号：WO02049006A1
公开(公告)日：2002-06-20
申请号：PCT/EP2001/014495
申请日：2001-12-10
IPC分类号： G10L15/07 , G10L15/22 , G10L15/26 , G10L15/06
CPC分类号： G10L15/07 , G10L15/26 , G10L2015/223
摘要： A mobile terminal (100) controllable by spoken utterances like proper names or keywords is described. The mobile terminal (100) comprises an interface (200) for receiving voice prompts, a model generator (430) for generating acoustic models based on the received voice prompts, and an automatic speech recognizer (110) for recognizing the spoken utterances based on the generated acoustic models.
摘要翻译：描述了可以通过诸如专有名称或关键字之类的语音话语来控制的移动终端（100）。移动终端（100）包括用于接收语音提示的接口（200），用于基于接收到的语音提示产生声学模型的模型发生器（430），以及用于基于所述语音提示识别所述口语话语的自动语音识别器（110）产生声学模型。

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