IP Library › Granted Patent US 10,090,005
Granted Patent B2
US 10,090,005 · App. 15/455,328 · Granted Oct 2, 2018

Analog voice activity detection

Inventors: Brandon David Rumberg (Morgantown, WV); David W. Graham (Morgantown, WV)
Assignee: ASPINITY, INC.
G10L25/84G10L15/02G10L15/142G10L21/038G10L25/18G10L25/21G10L25/78G10L25/81G10L2015/088G10L2025/783G10L2025/786
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,090,005
App. No.
15/455,328
Granted
Oct 2, 2018
Kind
B2
Abstract

According to some embodiments, an analog processing portion may receive an audio signal from a microphone. The analog processing portion may then convert the audio signal into sub-band signals and estimate an energy statistic value, such as a Signal-to-Noise Ratio (“SNR”) value, for each sub-band signal. A classification element may classify the estimated energy statistic values with analog processing such that a wakeup signal is generated when voice activity is detected. The wakeup signal may be associated with, for example, a battery-powered, always-listening audio application.

Claims (12)

1. A voice activity detection device, comprising:

an analog processing portion to receive an audio signal from a microphone, convert the audio signal into sub-band signals, and estimate an energy statistic value for each sub-band signal, including:

for each sub-band signal, a bandpass filter and one or more asymmetric integrators to output an energy value and a noise value for that sub-band signal; and

a classification element to classify the estimated energy statistic values with analog processing such that a wakeup signal is generated when voice activity is detected.

2. The device of claim 1 , wherein the estimated energy statistic value is a Signal-to-Noise Ratio (“SNR”) value.

3. The device of claim 1 , wherein the analog processing portion and classification element are integrated into the microphone.

4. The device of claim 1 , wherein the energy values and noise values are associated with feature extraction.

5. The device of claim 4 , wherein the feature extraction is further associated with transient noise features.

6. The device of claim 4 , wherein extracted features are output when the wakeup signal is generated.

7. The device of claim 4 , wherein extracted features are provided to a feature combination element.

8. The device of claim 7 , wherein the extracted features are provided to the feature combination element via a programmable switch matrix.

9. The device of claim 7 , wherein the feature combination element provides data to a decision integration element that generates the wakeup signal.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 10, 2017
From: RUMBERG, BRANDON DAVID; GRAHAM, DAVID W.
To: ASPINITY, INC.
Reel/Frame 041537/0004 →
Continuity (2)
Provisional Application 62306167 · Mar 10, 2016
Related Publication 20170263268A1 · Sep 14, 2017
Cited By (8)
US 12,327,182 US 12,327,183 US 12,347,421 US 12,393,833 US 12,579,420 US 12,579,421 US 12,718,079 US 12,748,960