-
公开(公告)号：US10789962B2
公开(公告)日：2020-09-29
申请号：US16186851
申请日：2018-11-12
Abstract: A system and method are presented for the correction of packet loss in audio in automatic speech recognition (ASR) systems. Packet loss correction, as presented herein, occurs at the recognition stage without modifying any of the acoustic models generated during training. The behavior of the ASR engine in the absence of packet loss is thus not altered. To accomplish this, the actual input signal may be rectified, the recognition scores may be normalized to account for signal errors, and a best-estimate method using information from previous frames and acoustic models may be used to replace the noisy signal.
-
Patent Agency Ranking
公开(公告)号：US11694697B2
公开(公告)日：2023-07-04
申请号：US16915160
申请日：2020-06-29
CPC classification number: G10L19/005 , G10L15/08 , G10L15/14 , G10L15/142 , G10L15/20 , G10L15/02 , G10L25/18 , G10L25/21 , G10L2015/025 , G10L2019/0012
Abstract: A system and method are presented for the correction of packet loss in audio in automatic speech recognition (ASR) systems. Packet loss correction, as presented herein, occurs at the recognition stage without modifying any of the acoustic models generated during training. The behavior of the ASR engine in the absence of packet loss is thus not altered. To accomplish this, the actual input signal may be rectified, the recognition scores may be normalized to account for signal errors, and a best-estimate method using information from previous frames and acoustic models may be used to replace the noisy signal.
-
公开(公告)号：US20190080701A1
公开(公告)日：2019-03-14
申请号：US16186851
申请日：2018-11-12
Abstract: A system and method are presented for the correction of packet loss in audio in automatic speech recognition (ASR) systems. Packet loss correction, as presented herein, occurs at the recognition stage without modifying any of the acoustic models generated during training. The behavior of the ASR engine in the absence of packet loss is thus not altered. To accomplish this, the actual input signal may be rectified, the recognition scores may be normalized to account for signal errors, and a best-estimate method using information from previous frames and acoustic models may be used to replace the noisy signal.
-
公开(公告)号：US11574642B2
公开(公告)日：2023-02-07
申请号：US16915160
申请日：2020-06-29
Abstract: A system and method are presented for the correction of packet loss in audio in automatic speech recognition (ASR) systems. Packet loss correction, as presented herein, occurs at the recognition stage without modifying any of the acoustic models generated during training. The behavior of the ASR engine in the absence of packet loss is thus not altered. To accomplish this, the actual input signal may be rectified, the recognition scores may be normalized to account for signal errors, and a best-estimate method using information from previous frames and acoustic models may be used to replace the noisy signal.
-
公开(公告)号：US11294955B2
公开(公告)日：2022-04-05
申请号：US16378452
申请日：2019-04-08
Inventor： Srinath Cheluvaraja , Ananth Nagaraja Iyer , Felix Immanuel Wyss
IPC: G06F16/683 , G10L25/51
Abstract: A system and method are presented for optimization of audio fingerprint search. In an embodiment, the audio fingerprints are organized into a recursive tree with different branches containing fingerprint sets that are dissimilar to each other. The tree is constructed using a clustering algorithm based on a similarity measure. The similarity measure may comprise a Hamming distance for a binary fingerprint or a Euclidean distance for continuous valued fingerprints. In another embodiment, each fingerprint is stored at a plurality of resolutions and clustering is performed hierarchically. The recognition of an incoming fingerprint begins from the root of the tree and proceeds down its branches until a match or mismatch is declared. In yet another embodiment, a fingerprint definition is generalized to include more detailed audio information than in the previous definition.
-
公开(公告)号：US20190236101A1
公开(公告)日：2019-08-01
申请号：US16378452
申请日：2019-04-08
Inventor： Srinath Cheluvaraja , Ananth Nagaraja Iyer , Felix Immanuel Wyss
IPC: G06F16/683 , G10L25/51
CPC classification number: G06F16/683 , G10L25/51
Abstract: A system and method are presented for optimization of audio fingerprint search. In an embodiment, the audio fingerprints are organized into a recursive tree with different branches containing fingerprint sets that are dissimilar to each other. The tree is constructed using a clustering algorithm based on a similarity measure. The similarity measure may comprise a Hamming distance for a binary fingerprint or a Euclidean distance for continuous valued fingerprints. In another embodiment, each fingerprint is stored at a plurality of resolutions and clustering is performed hierarchically. The recognition of an incoming fingerprint begins from the root of the tree and proceeds down its branches until a match or mismatch is declared. In yet another embodiment, a fingerprint definition is generalized to include more detailed audio information than in the previous definition.
-
-
-
-
-
Search Results
6
records
(took 0.011 seconds)
