US 12,308,046 B2
Inertial sensor unit and method for detecting a speech activity
Sergej Scheiermann, Reutlingen (DE); Omar Elrhoul, Reutlingen (DE); and Rainer Dorsch, Kirchentellinsfurt (DE)
Assigned to ROBERT BOSCH GMBH, Stuttgart (DE)
Filed by Robert Bosch GmbH, Stuttgart (DE)
Filed on Jun. 7, 2021, as Appl. No. 17/340,404.
Claims priority of application No. 102020208206.5 (DE), filed on Jul. 1, 2020.
Prior Publication US 2022/0005495 A1, Jan. 6, 2022
Int. Cl. G10L 25/78 (2013.01)
CPC G10L 25/78 (2013.01) 16 Claims
OG exemplary drawing
 
1. An inertial sensor unit, comprising:
a. a sensor element configured to detect and convert movements and vibrations into an electrical sensor signal;
b. a signal processing element configured to evaluate the sensor signal with a goal of detecting vibrations induced by a speech activity; and
c. an interface configured to signal a detected speech activity,
wherein
the signal processing element includes a first processing stage and a second processing stage for the sensor signal, the first processing stage being configured to check a first criterion for a presence of a speech activity, and the second processing stage configured to check at least one further, second criterion for the presence of a speech activity,
the second processing stage is passed through only if the sensor signal has passed through the first processing stage and the first criterion for the presence of a speech activity has been met, and
the signal processing element is configured to activate the interface for signaling a speech activity only if the sensor signal has passed through the second processing stage and the at least one further, second criterion for the presence of a speech activity has been met,
wherein the inertial sensor unit is configured to operate in different operating modes, in that individual components of the inertial sensor unit may be optionally activated or deactivated and/or operated in different operating modes,
wherein:
in a first operating mode, the first processing stage of the signal processing element is operated in the first operating mode and the second processing stage is deactivated;
in a second operating mode, the first processing stage of the signal processing element is operated in the second operating mode and the second processing stage is activated; and
the signal processing element is configured to automatically switch between the first operating mode and the second operating mode, depending on whether the first and/or the second criterion for the presence of a speech activity have been met,
wherein in the first operating mode:
a measurement occurs at low data rates by a low oversampling rate of the signal, and
the measurement occurs only via a single axis,
wherein in the second operating mode:
the measurement occurs at higher data rates by a higher oversampling rate of the signal, and
the measurement occurs via all axes.