Signal processing system and signal processing method
A signal processing and signal technology, used in signal processing, microphone signal combination, speech analysis, etc.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
no. 1 Embodiment approach
[0018] figure 1 It is a block diagram showing the configuration of the signal processing system 100-1 of the first embodiment. This signal processing system 100 - 1 includes a sensor unit 101 , a sound source separation unit 102 , a directional characteristic distribution calculation unit 103 , a similarity calculation unit 104 , and a connection unit 105 .
[0019] The sensor unit 101 receives a signal obtained by superimposing observation signals observed by a plurality of sensors. The above-mentioned sound source separation unit 102 estimates a separation matrix functioning as a filter for separating observation signals from signals received by the sensor unit 101 based on a frame unit based on a certain period of time, and extracts a signal from the received signal based on the separation matrix. A plurality of signals are separated, and each separated signal is output. The directional characteristic distribution calculating unit 103 calculates the directional characteri...
no. 2 Embodiment approach
[0089] image 3 It is a block diagram showing the configuration of the signal processing system 100-2 of the second embodiment. exist image 3 in, right with figure 1 The same parts are denoted by the same reference numerals, and repeated explanations are omitted.
[0090] The signal processing system 100-2 of the present embodiment is configured by adding a function that can further assign a relative positional relationship to the signals output by the first embodiment, and based on the configuration of the first embodiment, a direction estimation unit 106 is added. and the positional relationship imparting unit 107 .
[0091] The direction estimating unit 106 determines the positional relationship of the separated signals based on the separation matrix obtained by the sound source separating unit 102 . That is, the directional characteristic distribution corresponding to the k-th separated signal is represented by the following equation (16).
[0092]
[0093] Here, ...
no. 3 Embodiment approach
[0101] Figure 4 It is a block diagram showing the configuration of the signal processing system 100-3 of the third embodiment. exist Figure 4 in, right with figure 1 The same parts are denoted by the same reference numerals, and repeated explanations are omitted.
[0102] In the present embodiment, by introducing a voice interval detection (VAD; Voice Activity Detection) method to the first embodiment or its derivatives, it is prevented that the cumulative distribution is updated to an undesired distribution due to noise other than the target voice. Specifically, as Figure 4 As shown, for each separated signal of a plurality of separated signals obtained by the sound source separation unit 102, it is determined by the sound interval detection unit 109 which one is the sound interval or the non-sound interval, and only the channel corresponding to the sound interval is considered The cumulative distribution of is updated by the similarity calculation unit 104, and the cu...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More