Unlock instant, AI-driven research and patent intelligence for your innovation.

Wake on voice key phrase segmentation

A key phrase, speech technology, used in speech analysis, speech recognition, instruments, etc.

Pending Publication Date: 2019-11-15
INTEL CORP
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, a large number of important questions remain for keyphrase segmentation techniques

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Wake on voice key phrase segmentation
  • Wake on voice key phrase segmentation
  • Wake on voice key phrase segmentation

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0085] Example 1 is a method for keyphrase segmentation comprising: generating a set of acoustic scores by a neural network based on the accumulation of feature vectors extracted from temporal segments of an audio signal, the set of Each acoustic score in the acoustic score represents the likelihood of a class of speech associated with a time segment; a progression of a sequence of scoring model states is generated by the keyphrase model decoder, each based on a sequence of time-segments from the audio signal. the detection of a phonetic unit associated with a corresponding set of acoustic scores in the set of acoustic scores generated by the segment; the progression of the sequence of scoring states is analyzed by the keyphrase segmentation circuit to detect a pattern associated with the progression; and the progression by the keyphrase segmentation circuit Starting and ending points for segmentation of key phrases are determined based on the alignment of the detected pattern ...

example 2

[0086] Example 2 includes the subject matter of Example 1 and further includes detecting key phrases based on the accumulation and propagation of the acoustic scores in the acoustic score set.

example 3

[0087] Example 3 includes the subject matter of Examples 1 or 2, wherein the determination of the onset is further based on one of the time segments associated with the detection of the key phrase.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The disclosure relates to wake on voice key phrase segmentation. Techniques are provided for segmentation of a key phrase. A methodology implementing the techniques according to an embodiment includesaccumulating feature vectors extracted from time segments of an audio signal, and generating a set of acoustic scores based on those feature vectors. Each of the acoustic scores in the set representsa probability for a phonetic class associated with the time segments. The method further includes generating a progression of scored model state sequences, each of the scored model state sequences based on detection of phonetic units associated with a corresponding one of the sets of acoustic scores generated from the time segments of the audio signal. The method further includes analyzing the progression of scored state sequences to detect a pattern associated with the progression, and determining a starting and ending point for segmentation of the key phrase based on alignment of the detected pattern with an expected pattern.

Description

technical field [0001] This disclosure relates to wake-up speech keyphrase segmentation. Background technique [0002] Key phrase detection is an important feature in voice-enabled devices. The device can be woken from a low power listening state by the utterance of a specific key phrase from the user. Keyphrases detect the event initiator's conversation with the device, for example, a command or question to a personal assistant. This dialog includes further processing of the user's speech, and the effectiveness of this processing depends largely on the accuracy with which the boundaries of key phrases in the audio signal are determined (this process is called key phrase segmentation). However, a large number of important questions remain for keyphrase segmentation techniques. Contents of the invention [0003] According to an aspect of the present disclosure, there is provided a method for keyphrase segmentation, the method comprising: generating a set of acoustic scor...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/05G10L15/14G10L15/16G10L15/22G10L15/02
CPCG10L15/02G10L15/05G10L15/142G10L15/16G10L15/22G10L15/04G10L2015/025
Inventor 托马什·多劳托比亚斯·博克雷普热米司勒·托马谢夫斯基塞巴斯蒂安·切里巴尤利乌斯·诺尔曼·霍耶茨基
Owner INTEL CORP