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

US20190156819A1 COMPLEX EVOLUTION RECURRENT NEURAL NETWORKS 审中-公开
公开(公告)号：US20190156819A1
公开(公告)日：2019-05-23
申请号：US16251430
申请日：2019-01-18
申请人： Google LLC
发明人： Izhak Shafran , Thomas E. Bagby , Russell John Wyatt Skerry-Ryan
IPC分类号： G10L15/16 , G10L19/02 , G10L15/02 , G10H1/00
CPC分类号： G10L15/16 , G06N3/02 , G10H1/00 , G10H2210/036 , G10H2210/046 , G10H2250/235 , G10H2250/311 , G10L15/02 , G10L17/18 , G10L19/0212 , G10L25/30
摘要： Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for speech recognition using complex evolution recurrent neural networks. In some implementations, audio data indicating acoustic characteristics of an utterance is received. A first vector sequence comprising audio features determined from the audio data is generated. A second vector sequence is generated, as output of a first recurrent neural network in response to receiving the first vector sequence as input, where the first recurrent neural network has a transition matrix that implements a cascade of linear operators comprising (i) first linear operators that are complex-valued and unitary, and (ii) one or more second linear operators that are non-unitary. An output vector sequence of a second recurrent neural network is generated. A transcription for the utterance is generated based on the output vector sequence generated by the second recurrent neural network. The transcription for the utterance is provided.

2. 发明申请

US20190057678A1 Means and methods for switching odd and even numbers of matched pickups to produce all humbucking tones 审中-公开
公开(公告)号：US20190057678A1
公开(公告)日：2019-02-21
申请号：US16139027
申请日：2018-09-22
申请人： Donald L Baker
发明人： Donald L Baker
IPC分类号： G10H3/18 , G10H1/46 , G10H1/18 , G10H1/34 , G10H3/22
CPC分类号： G10H3/181 , G10H1/06 , G10H1/18 , G10H1/342 , G10H1/46 , G10H3/182 , G10H3/185 , G10H3/186 , G10H3/188 , G10H3/22 , G10H2220/505 , G10H2250/235
摘要： This invention discloses a switching system for any odd or even number of two or more matched vibrations sensors, such that all possible circuits of such sensors that can be produced by the system are humbucking, rejecting external interferences signals. The sensors must be matched, especially with respect to response to external hum and internal impedance, and be capable of being made or arranged so that the responses of individual sensors to vibration can be inverted, compared to another matched sensor, placed in the same physical position, while the interference signal is not. Such that for 2, 3, 4, 5, 6, 7 and 8 sensors, there exist 1, 6, 25, 90, 301, 966 and 3025 unique humbucking circuits, respectively, with signal outputs that can be either single-ended or differential. Embodiments of switching systems include electro-mechanical switches, programmable switches, solid-state digital-analog switches, and micro-controller driven solid state switches using time-series to spectral-series transforms to pick the order of tones from bright to warm and back.

3. 发明申请

US20170372724A1 ANALYZING CHANGES IN VOCAL POWER WITHIN MUSIC CONTENT USING FREQUENCY SPECTRUMS 有权
公开(公告)号：US20170372724A1
公开(公告)日：2017-12-28
申请号：US15331651
申请日：2016-10-21
申请人： Microsoft Technology Licensing, LLC
发明人： Stewart Paul Tootill , Kevin Lingley , David Niall Coghlan , Michal Vine , Linden Vongsathorn
IPC分类号： G10L25/18 , G10L21/0308 , G10L25/27 , G10L25/51
CPC分类号： G10L25/18 , G10H1/00 , G10H1/125 , G10H2210/051 , G10H2210/061 , G10H2250/235 , G10H2250/455 , G10L21/028 , G10L21/0308 , G10L25/27 , G10L25/51
摘要： Technologies are described for identifying familiar or interesting parts of music content by analyzing changes in vocal power using frequency spectrums. For example, a frequency spectrum can be generated from digitized audio. Using the frequency spectrum, the harmonic content and percussive content can be separated. The vocal content can then be separated from the harmonic and/or percussive content. The vocal content can then be processed to identify surge points in the digitized audio. In some implementations, the vocal content is included in the harmonic content during the separation procedure and is then separated from the harmonic content.

4. 发明授权

US09666199B2 Automatic conversion of speech into song, rap, or other audible expression having target meter or rhythm 有权
公开(公告)号：US09666199B2
公开(公告)日：2017-05-30
申请号：US13910949
申请日：2013-06-05
申请人： Smule, Inc.
发明人： Parag Chordia , Mark Godfrey , Alexander Rae , Prerna Gupta , Perry R. Cook
IPC分类号： G10L21/007 , G10L21/04 , G10L19/02 , G10L19/00 , G10L21/055 , G10H1/36
CPC分类号： G10L19/02 , G10H1/366 , G10H2210/051 , G10H2240/141 , G10H2250/235 , G10L19/00 , G10L21/055
摘要： Captured vocals may be automatically transformed using advanced digital signal processing techniques that provide captivating applications, and even purpose-built devices, in which mere novice user-musicians may generate, audibly render and share musical performances. In some cases, the automated transformations allow spoken vocals to be segmented, arranged, temporally aligned with a target rhythm, meter or accompanying backing tracks and pitch corrected in accord with a score or note sequence. Speech-to-song music applications are one such example. In some cases, spoken vocals may be transformed in accord with musical genres such as rap using automated segmentation and temporal alignment techniques, often without pitch correction. Such applications, which may employ different signal processing and different automated transformations, may nonetheless be understood as speech-to-rap variations on the theme.

5. 发明授权

US09553553B2 Engine sound synthesis system 有权
标题翻译：发动机声音综合系统
公开(公告)号：US09553553B2
公开(公告)日：2017-01-24
申请号：US13940884
申请日：2013-07-12
申请人： Harman Becker Automotive Systems GmbH
发明人： Markus Christoph
IPC分类号： H03G3/00 , G10H1/08 , G10K15/02 , G10L19/06
CPC分类号： H03G3/00 , G10H1/08 , G10H2210/066 , G10H2220/211 , G10H2250/235 , G10H2250/381 , G10H2250/471 , G10K15/02 , G10L19/06
摘要： An engine sound synthesis system is operable to analyze sound. Operation of the system may include providing an input sound signal to be analysed and determining a fundamental frequency of the input signal from the input signal or from at least one guide signal. Furthermore, the frequencies of higher harmonics of the fundamental frequency are determined, thus determining harmonic model parameters. A harmonic signal based on the harmonic model parameters is synthesized and a residual signal is estimated by subtracting the harmonic signal from the input signal. Residual model parameters are estimated based on the residual signal. Furthermore, a corresponding method for synthesizing a sound signal is described.
摘要翻译：引擎声音合成系统可操作以分析声音。系统的操作可以包括提供要分析的输入声音信号，并根据输入信号或至少一个引导信号确定输入信号的基频。此外，确定基频的高次谐波频率，从而确定谐波模型参数。合成基于谐波模型参数的谐波信号，并通过从输入信号中减去谐波信号来估计残差信号。剩余模型参数根据剩余信号估计。此外，描述了用于合成声音信号的相应方法。

6. 发明申请

US20160112225A1 Measuring Waveforms With The Digital Infinite Exponential Transform 审中-公开
标题翻译：数字无限指数变换测量波形
公开(公告)号：US20160112225A1
公开(公告)日：2016-04-21
申请号：US14885693
申请日：2015-10-16
申请人： Digital Harmonic LLC
发明人： Frederick M. SLAY
IPC分类号： H04L27/156
CPC分类号： H04L27/156 , G01R23/16 , G06F17/14 , G10H2210/056 , G10H2250/215 , G10H2250/235 , G10H2250/251 , G10H2250/261 , G10L25/18 , G10L25/48
摘要： A method is performed of selecting a signal of interest from a compound signal. The method includes generating a matrix and initializing a first row of the matrix to zero. The compound signal is obtained as a digital waveform signal, with sampling rate S samples per second. For each sample of the digital signal, a new entry is recursively computed for the matrix. For each frequency bin in the matrix (where f is the center frequency of the bin) the value in the new row is computed by multiplying the value for that bin in the previous row by the complex number r*ei2πf/S and adding the new signal sample multiplied by a real constant. The method includes identifying the signal of interest from the matrix, whereby an uncertainty of which frequency bin the signal of interest exists in is eliminated.
摘要翻译：执行从复合信号中选择感兴趣的信号的方法。该方法包括生成矩阵并将矩阵的第一行初始化为零。获得复合信号作为数字波形信号，采样速率S采样每秒。对于数字信号的每个样本，对矩阵递归计算一个新的条目。对于矩阵中的每个频率仓（其中f是仓的中心频率），新行中的值通过将上一行中的该仓的值乘以复数r * ei2＆pgr; f / S并添加新信号样本乘以实常数。该方法包括从矩阵中识别感兴趣的信号，由此消除了感兴趣的信号存在哪个频率bin的不确定性。

7. 发明授权

US09313359B1 Media content identification on mobile devices 有权
标题翻译：移动设备上的媒体内容标识
公开(公告)号：US09313359B1
公开(公告)日：2016-04-12
申请号：US13590701
申请日：2012-08-21
申请人： Mihailo M. Stojancic , Jose Pio Pereira , Peter Wendt , Shashank Merchant , Sunil Suresh Kulkarni
发明人： Mihailo M. Stojancic , Jose Pio Pereira , Peter Wendt , Shashank Merchant , Sunil Suresh Kulkarni
IPC分类号： H04N1/32 , H04N5/44 , H04N21/439
CPC分类号： H04N21/4394 , G10H1/0008 , G10H2210/041 , G10H2210/051 , G10H2240/141 , G10H2250/235 , H04N1/32 , H04N21/41407 , H04N21/42203 , H04N21/4223 , H04N21/4332 , H04N21/439 , H04N21/8352 , H04N21/84 , H04N21/8456
摘要： A mobile device responds in real time to media content presented on a media device, such as a television. The mobile device captures temporal fragments of audio-video content on its microphone, camera, or both and generates corresponding audio-video query fingerprints. The query fingerprints are transmitted to a search server located remotely or used with a search function on the mobile device for content search and identification. Audio features are extracted and audio signal global onset detection is used for input audio frame alignment. Additional audio feature signatures are generated from local audio frame onsets, audio frame frequency domain entropy, and maximum change in the spectral coefficients. Video frames are analyzed to find a television screen in the frames, and a detected active television quadrilateral is used to generate video fingerprints to be combined with audio fingerprints for more reliable content identification.
摘要翻译：移动设备实时响应媒体设备（如电视机）上呈现的媒体内容。移动设备捕获其麦克风，相机或两者上的音频 - 视频内容的时间片段，并生成相应的音视频查询指纹。查询指纹被发送到位于远程的搜索服务器或与移动设备上的搜索功能一起使用，用于内容搜索和识别。提取音频特征，音频信号全局起始检测用于输入音频帧对齐。额外的音频特征签名由本地音频帧发送，音频帧频域熵和频谱系数的最大变化产生。视频帧被分析以在帧中找到电视屏幕，并且使用检测到的活动电视四边形来生成视频指纹以与音频指纹组合以用于更可靠的内容识别。

8. 发明申请

US20140180674A1 AUDIO MATCHING WITH SEMANTIC AUDIO RECOGNITION AND REPORT GENERATION 有权
标题翻译：音频与语音识别和报告生成的匹配
公开(公告)号：US20140180674A1
公开(公告)日：2014-06-26
申请号：US13725004
申请日：2012-12-21
申请人： ARBITRON INC.
发明人： Alan Neuhauser , John Stavropoulos
IPC分类号： G10L19/018
CPC分类号： G10H1/0008 , G10H2210/036 , G10H2210/041 , G10H2210/056 , G10H2210/066 , G10H2210/071 , G10H2210/076 , G10H2210/081 , G10H2250/235 , G10H2250/251 , G10L15/00 , G10L25/51 , G10L25/81 , G10L25/84
摘要： System, apparatus and method for determining semantic information from audio, where incoming audio is sampled and processed to extract audio features, including temporal, spectral, harmonic and rhythmic features. The extracted audio features are compared to stored audio templates that include ranges and/or values for certain features and are tagged for specific ranges and/or values. The semantic information may be associated with audio signature dataExtracted audio features that are most similar to one or more templates from the comparison are identified according to the tagged information. The tags are used to determine the semantic audio data that includes genre, instrumentation, style, acoustical dynamics, and emotive descriptor for the audio signal.
摘要翻译：用于从音频确定语义信息的系统，装置和方法，其中对输入音频进行采样和处理以提取音频特征，包括时间，频谱，谐波和节奏特征。所提取的音频特征与存储的音频模板进行比较，所述音频模板包括特定特征的范围和/或值，并针对特定范围和/或值进行标记。语义信息可以与音频签名数据相关联。根据标记的信息来识别与来自比较的一个或多个模板最相似的提取的音频特征。这些标签用于确定语音音频数据，包括音频信号的类型，仪器，风格，声学动力学和情感描述符。

9. 发明授权

US08618402B2 Musical harmony generation from polyphonic audio signals 有权
标题翻译：来自复音音频信号的音乐和谐生成
公开(公告)号：US08618402B2
公开(公告)日：2013-12-31
申请号：US13646366
申请日：2012-10-05
申请人： Glen A. Rutledge , William Norman Campbell , Peter R. Lupini
发明人： Glen A. Rutledge , William Norman Campbell , Peter R. Lupini
IPC分类号： G10H1/00
CPC分类号： G10H1/36 , G10H1/38 , G10H3/125 , G10H2210/261 , G10H2250/235
摘要： Melody and accompaniment audio signals are received and processed to identify one or more harmony notes and a harmony signal is produced based on the one or more harmony notes. Typically the melody note is identified and a spectrum of the accompaniment audio signal and is obtained, and one or more harmony notes are identified based on the melody note and the accompaniment spectrum. The melody, and accompaniment signals can be processed in real-time for combination with the harmony signal in an audio performance. In some examples, audio signals are processed and harmonies generated for subsequent performance based on, for example, MIDI files generated from the audio signals.
摘要翻译：接收和处理旋律和伴奏音频信号以识别一个或多个和声音符，并且基于一个或多个和声音符产生和声信号。通常，确定旋律音符并获得伴奏音频信号的频谱，并且基于旋律音符和伴奏频谱来识别一个或多个和声音符。旋律和伴奏信号可以实时处理，与音乐表演中的和声信号相结合。在一些示例中，基于例如从音频信号产生的MIDI文件，对音频信号进行处理并产生用于后续演出的和声。

10. 发明授权

US08442816B2 Music-piece classification based on sustain regions 有权
标题翻译：基于维持区域的音乐片段分类
公开(公告)号：US08442816B2
公开(公告)日：2013-05-14
申请号：US12929713
申请日：2011-02-10
申请人： Ichiro Shishido
发明人： Ichiro Shishido
IPC分类号： G10H2240/075 , G10L11/00 , G10L19/02
CPC分类号： G10H1/0008 , G10H2210/066 , G10H2240/081 , G10H2250/235 , G10L25/90
摘要： Audio data representative of a music piece is converted into data components in respective different frequency bands for every unit time interval to generate time frequency data pieces assigned to the respective different frequency bands. From the generated time frequency data pieces, detection is made as to each sustain region in which an effective data component in one of the frequency bands continues to occur during a reference time interval or longer. A feature quantity is calculated from at least one of (1) a number of the detected sustain regions and (2) magnitudes of the effective data components in the detected sustain regions. The music piece is classified in response to the calculated feature quantity.
摘要翻译：代表乐曲的音频数据在每个单位时间间隔被转换成各个不同频带的数据分量，以产生分配给各个不同频带的时间频率数据。从所生成的时间频率数据中，对参考时间间隔或更长时间内持续发生频带中的一个频带中的有效数据分量的每个维持区域进行检测。从（1）检测到的维持区域的数量和（2）检测到的维持区域中的有效数据分量的大小中的至少一个计算特征量。乐曲根据计算的特征量进行分类。

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