公开(公告)号：US20130254422A2

公开(公告)日：2013-09-26

申请号：US12773753

申请日：2010-05-04

Applicant: Aaron Steven Master ,  Timothy P. Stonehocker ,  Benjamin John Levitt ,  Jun Huang ,  Keyvan Mohajer

Inventor： Aaron Steven Master ,  Timothy P. Stonehocker ,  Benjamin John Levitt ,  Jun Huang ,  Keyvan Mohajer

IPC: G06F17/30 ,  G06F15/16 ,  H04B15/00

CPC classification number: G06F17/30743 ,  G06F17/30749 ,  G06F17/30758 ,  G10L25/51

Abstract: Systems and methods for recognizing sounds are provided herein. User input relating to one or more sounds is received from a computing device. Instructions, which are stored in memory, are executed by a processor to discriminate the one or more sounds, extract music features from the one or more sounds, analyze the music features using one or more databases, and obtain information regarding the music features based on the analysis. Further, information regarding the music features of the one or more sounds may be transmitted to display on the computing device.

Abstract translation: 本文提供用于识别声音的系统和方法。 从计算设备接收与一个或多个声音相关的用户输入。 存储在存储器中的指令由处理器执行以区分一个或多个声音，从一个或多个声音中提取音乐特征，使用一个或多个数据库分析音乐特征，并且基于以下方式获得关于音乐特征的信息： 分析。 此外，可以发送关于一个或多个声音的音乐特征的信息以在计算设备上显示。

公开(公告)号：US08433431B1

公开(公告)日：2013-04-30

申请号：US13372399

申请日：2012-02-13

Applicant: Aaron Master ,  Kamyar Mohajer ,  Keyvan Mohajer ,  Timothy P. Stonehocker ,  Bernard Mont-Reynaud

Inventor： Aaron Master ,  Kamyar Mohajer ,  Keyvan Mohajer ,  Timothy P. Stonehocker ,  Bernard Mont-Reynaud

IPC: G06F17/00

CPC classification number: G10H1/0008 ,  G10H2220/011 ,  G10H2240/325 ,  G10L25/54 ,  H04M2203/305

Abstract: Technologies described relate to coordination between audio and tracking of the corresponding text in an audio recognition mode and an audio playback mode. Optionally, audio recognition includes receiving a signal corresponding to an audio track; determining an identity of the audio track and a current audio position within the audio track; displaying on a display a portion of a text that is linked and synchronized to the audio track identified and automatically scrolling the portion of the text displayed in pace with the audio track playing; and displaying a current text indicator that emphasizes current text, wherein the current text indicator is visually synchronized on the display to current audio playing from the audio track. Optionally include redetermining the current audio position in the audio track from the signal received and updating synchronization of the current text indicator with the current audio playing.

Abstract translation: 所描述的技术涉及在音频识别模式和音频回放模式中的对应文本的音频和跟踪之间的协调。 可选地，音频识别包括接收对应于音频轨道的信号; 确定音轨的身份和音轨内的当前音频位置; 在显示器上显示与所识别的音频轨道链接并同步的文本的一部分，并且自动地滚动与音轨播放一起显示的文本的部分; 以及显示强调当前文本的当前文本指示符，其中当前文本指示符在显示器上可视地同步到来自音频轨道的当前音频播放。 可选地包括从接收到的信号重新确定音频轨道中的当前音频位置，并且更新当前文本指示符与当前音频播放的同步。

公开(公告)号：US20100121643A1

公开(公告)日：2010-05-13

申请号：US12610920

申请日：2009-11-02

Applicant: Keyvan Mohajer ,  Seyed Majid Emami ,  Jon Grossman ,  Joe Kyaw Soe Aung ,  Sina Sohangir

Inventor： Keyvan Mohajer ,  Seyed Majid Emami ,  Jon Grossman ,  Joe Kyaw Soe Aung ,  Sina Sohangir

IPC: G10L15/04 ,  G10L15/18 ,  G10L15/14

CPC classification number: G10L15/142 ,  G10L15/05 ,  G10L15/32

Abstract: The technology disclosed relates to a system and method for fast, accurate and parallelizable speech search, called Crystal Decoder. It is particularly useful for search applications, as opposed to dictation. It can achieve both speed and accuracy, without sacrificing one for the other. It can search different variations of records in the reference database without a significant increase in elapsed processing time. Even the main decoding part can be parallelized as the number of words increase to maintain a fast response time.

Abstract translation: 所公开的技术涉及一种称为晶体解码器的用于快速，准确和可并行化语音搜索的系统和方法。 这对于搜索应用程序来说特别有用，而不是听写。 它可以实现速度和精度，而不会牺牲另一个。 它可以搜索参考数据库中不同的记录变体，而不会显着增加经过的处理时间。 即使主解码部分可以并行化，因为字数增加以保持快速的响应时间。

公开(公告)号：US07698254B2

公开(公告)日：2010-04-13

申请号：US11778752

申请日：2007-07-17

Applicant: Michal Grabowski ,  Majid Emami ,  James M. Hom ,  Keyvan Mohajer

Inventor： Michal Grabowski ,  Majid Emami ,  James M. Hom ,  Keyvan Mohajer

IPC: G06F17/30 ,  G06F17/00

CPC classification number: G06F17/30693 ,  G06F17/30646 ,  G06F17/30681 ,  Y10S707/99932

Abstract: A method for searching a database to produce search results from queries likely to contain errors. The process begins by identifying database features likely to be useful in searching, and those features are employed to index the database. After receiving a query from a user, the system develops a rough score for the query, by extracting features from the query, assigning match scores to query features matching database features; and assigning approximation scores to query features amenable to approximation analysis with database features. The rough score is used to identify identifying a set of database records for further analysis. Those records are then subjected to a more detailed rescoring process, based on correspondence between individual query elements and individual record elements, and between the query and the database record content, taken as a whole. Based on the rescoring process, output is provided to the user.

Abstract translation: 一种用于搜索数据库以从可能包含错误的查询中产生搜索结果的方法。 该过程开始于识别可能在搜索中有用的数据库特征，并且使用这些特征来索引数据库。 在从用户接收到查询后，系统通过从查询中提取特征，将匹配分数分配给查询匹配数据库特征的特征，为查询开发粗略的分数; 并将近似分数分配给具有数据库特征的近似分析的查询特征。 粗略分数用于识别识别一组数据库记录以供进一步分析。 然后，根据个别查询元素和各个记录元素之间的对应关系，以及作为整体的查询和数据库记录内容之间，对这些记录进行更详细的记录过程。 基于挽救过程，向用户提供输出。

公开(公告)号：US20100030775A1

公开(公告)日：2010-02-04

申请号：US12578531

申请日：2009-10-13

Applicant: Keyvan Mohajer ,  Majid Emami ,  Michal Grabowski ,  James M. Hom

Inventor： Keyvan Mohajer ,  Majid Emami ,  Michal Grabowski ,  James M. Hom

IPC: G06F17/30 ,  G06F17/00 ,  G10L15/06

CPC classification number: G06F17/30758 ,  G06F17/30743 ,  G06F17/30749 ,  Y10S707/913 ,  Y10S707/916 ,  Y10S707/99931 ,  Y10S707/99933 ,  Y10S707/99935 ,  Y10S707/99943 ,  Y10S707/99948

Abstract: A method for non-text-based identification of a selected item of stored music. The first broad portion of the method focuses on building a music identification database. That process requires capturing a tag of the selected musical item, and processing the tag to develop reference key to the same. Then the tag is stored, together with the reference key and an association to the stored music. The database is built by collecting a multiplicity of tags. The second broad portion of the method is retrieving a desired item of stored music from the database. That process calls for capturing a query tag from a user, and processing the query tag to develop a query key to the same. The query tag is compared to reference keys stored in the database to identify the desired item of stored music.

Abstract translation: 一种用于存储音乐的所选项目的非基于文本的识别的方法。 该方法的第一大部分重点是建立一个音乐识别数据库。 该过程需要捕获所选择的音乐作品的标签，并处理标签以将参考键打开。 然后，标签与参考键一起存储，并与存储的音乐相关联。 数据库是通过收集多个标签来构建的。 方法的第二大部分是从数据库中检索所需的存储音乐项目。 该过程需要从用户捕获查询标签，并处理查询标签以将查询密钥设置为相同。 将查询标签与存储在数据库中的参考键进行比较，以识别所需的存储音乐项。

公开(公告)号：US20080167862A1

公开(公告)日：2008-07-10

申请号：US11971070

申请日：2008-01-08

Applicant: Keyvan Mohajer

Inventor： Keyvan Mohajer

IPC: G10L11/04

CPC classification number: G10L25/90 ,  G10L15/063

Abstract: A method for employing pitch in a speech recognition engine. The process begins by building training models of selected speech samples, a process which begins by analyzing each sample as a sequential series of frames, each frame having a selected duration and overlap with adjacent frames. A pitch estimate of each frame is detected and recorded, and the pitch data is normalized, and the speech recognition parameters of the model are determined, after which the model is stored. Models are stored and updated for each of the set of training samples. The system is then employed to recognizing the speech content of a subject, which begins by analyzing the subject as a sequential series of frames, each frame having a selected duration and overlap with adjacent frames. A pitch estimate for each frame is detected and recorded, and the pitch data is normalized. Speech recognition techniques are then employed to recognize the content of the subject, employing the stored models.

Abstract translation: 一种在语音识别引擎中采用音调的方法。 该过程开始于构建所选语音样本的训练模型，该过程通过将每个样本分析为连续的一系列帧开始，每个帧具有选定的持续时间并与相邻帧重叠。 检测并记录每帧的音调估计，并且对音调数据进行归一化，并且确定模型的语音识别参数，然后存储模型。 为每组训练样本中的每一组存储和更新模型。 然后，该系统用于识别被摄体的语音内容，其通过将对象分析为连续的一系列帧开始，每帧具有选定的持续时间并与相邻帧重叠。 检测并记录每帧的音调估计，并且对音调数据进行归一化。 然后使用语音识别技术来识别被摄体的内容，采用所存储的模型。

公开(公告)号：US09390167B2

公开(公告)日：2016-07-12

申请号：US13193514

申请日：2011-07-28

Applicant: Bernard Mont-Reynaud ,  Aaron Master ,  Timothy P. Stonehocker ,  Keyvan Mohajer

Inventor： Bernard Mont-Reynaud ,  Aaron Master ,  Timothy P. Stonehocker ,  Keyvan Mohajer

IPC: G06F17/00 ,  G06F17/30

CPC classification number: G06F17/30743 ,  G06F17/30026 ,  G06F17/30749 ,  G06F17/30772

Abstract: The present invention relates to the continuous monitoring of an audio signal and identification of audio items within an audio signal. The technology disclosed utilizes predictive caching of fingerprints to improve efficiency. Fingerprints are cached for tracking an audio signal with known alignment and for watching an audio signal without known alignment, based on already identified fingerprints extracted from the audio signal. Software running on a smart phone or other battery-powered device cooperates with software running on an audio identification server.

Abstract translation: 本发明涉及音频信号的连续监视和音频信号内的音频项目的识别。 所公开的技术利用指纹的预测性缓存来提高效率。 基于从音频信号提取的已经识别的指纹，缓存指纹用于跟踪具有已知对准的音频信号并且用于观看没有已知对准的音频信号。 在智能手机或其他电池供电设备上运行的软件与在音频识别服务器上运行的软件配合使用。

公开(公告)号：US09280598B2

公开(公告)日：2016-03-08

申请号：US12773753

申请日：2010-05-04

Applicant: Aaron Steven Master ,  Timothy P. Stonehocker ,  Benjamin John Levitt ,  Jun Huang ,  Keyvan Mohajer

Inventor： Aaron Steven Master ,  Timothy P. Stonehocker ,  Benjamin John Levitt ,  Jun Huang ,  Keyvan Mohajer

IPC: G06F17/00 ,  G06F17/30

CPC classification number: G06F17/30743 ,  G06F17/30749 ,  G06F17/30758 ,  G10L25/51

Abstract: Systems and methods for recognizing sounds are provided herein. User input relating to one or more sounds is received from a computing device. Instructions, which are stored in memory, are executed by a processor to discriminate the one or more sounds, extract music features from the one or more sounds, analyze the music features using one or more databases, and obtain information regarding the music features based on the analysis. Further, information regarding the music features of the one or more sounds may be transmitted to display on the computing device.

Abstract translation: 本文提供用于识别声音的系统和方法。 从计算设备接收与一个或多个声音相关的用户输入。 存储在存储器中的指令由处理器执行以区分一个或多个声音，从一个或多个声音中提取音乐特征，使用一个或多个数据库分析音乐特征，并且基于以下方式获得关于音乐特征的信息： 分析。 此外，可以发送关于一个或多个声音的音乐特征的信息以在计算设备上显示。

公开(公告)号：US09196242B1

公开(公告)日：2015-11-24

申请号：US13482792

申请日：2012-05-29

Applicant: Aaron Master ,  Bernard Mont-Reynaud ,  Keyvan Mohajer

Inventor： Aaron Master ,  Bernard Mont-Reynaud ,  Keyvan Mohajer

IPC: G06F17/30 ,  G10L15/08

CPC classification number: G06F17/30743 ,  G10L15/08 ,  G10L25/54

Abstract: In one implementation, a method is described of retrying matching of an audio query against audio references. The method includes receiving a follow-up query that requests a retry at matching a previously submitted audio query. In some implementations, this follow-up query is received without any recognition hint that suggests how to retry matching. The follow-up query includes the audio query or a reference to the audio query to be used in the retry. The method further includes retrying matching the audio query using retry matching resources that include an expanded group of audio references, identifying at least one match and transmitting a report of the match. Optionally, the method includes storing data that correlates the follow-up query, the audio query or the reference to the audio query, and the match after retrying.

Abstract translation: 在一个实现中，描述了一种方法来重复音频查询与音频引用的匹配。 该方法包括接收在先前提交的音频查询匹配时请求重试的后续查询。 在一些实现中，接收到该后续查询，而没有提示如何重试匹配的任何识别提示。 后续查询包括音频查询或对要在重试中使用的音频查询的引用。 该方法还包括使用包括扩展的音频参考组的重试匹配资源来重试匹配音频查询，识别至少一个匹配并发送匹配的报告。 可选地，该方法包括存储将后续查询，音频查询或对音频查询的引用与重试后的匹配相关联的数据。

公开(公告)号：US20120036156A1

公开(公告)日：2012-02-09

申请号：US13275193

申请日：2011-10-17

Applicant: Keyvan Mohajer ,  Majid Emami ,  Michal Grabowski ,  James M. Hom

Inventor： Keyvan Mohajer ,  Majid Emami ,  Michal Grabowski ,  James M. Hom

IPC: G06F17/30

CPC classification number: G06F17/30758 ,  G06F17/30743 ,  G06F17/30749 ,  Y10S707/913 ,  Y10S707/916 ,  Y10S707/99931 ,  Y10S707/99933 ,  Y10S707/99935 ,  Y10S707/99943 ,  Y10S707/99948

Abstract: A method for non-text-based identification of a selected item of stored music. The first broad portion of the method focuses on building a music identification database. That process requires capturing a tag of the selected musical item, and processing the tag to develop reference key to the same. Then the tag is stored, together with the reference key and an association to the stored music. The database is built by collecting a multiplicity of tags. The second broad portion of the method is retrieving a desired item of stored music from the database. That process calls for capturing a query tag from a user, and processing the query tag to develop a query key to the same. The query tag is compared to reference keys stored in the database to identify the desired item of stored music.

Abstract translation: 一种用于存储音乐的所选项目的非基于文本的识别的方法。 该方法的第一大部分重点是建立一个音乐识别数据库。 该过程需要捕获所选择的音乐作品的标签，并处理标签以将参考键打开。 然后，标签与参考键一起存储，并与存储的音乐相关联。 数据库是通过收集多个标签来构建的。 方法的第二大部分是从数据库中检索所需的存储音乐项目。 该过程需要从用户捕获查询标签，并处理查询标签以将查询密钥设置为相同。 将查询标签与存储在数据库中的参考键进行比较，以识别所需的存储音乐项。