Primary transmission site switching in a multipoint videoconference environment based on human voice
First Claim
Patent Images
1. A method for determining a talk/listen state using voice detection, comprising:
- receiving an audio sample representing sound measured during a sample time interval;
detecting whether the audio sample includes voiced sound;
deriving an audio level from the audio sample, the audio level representing an average power level of the audio sample;
comparing the audio level to a threshold level;
determining the talk/listen state depending on a relation of the audio level to the threshold level and depending on whether the audio sample includes voiced sound.
6 Assignments
0 Petitions
Accused Products
Abstract
A method for determining a talk/listen state using voice detection includes receiving an audio sample and detecting whether the audio sample includes voiced sound. The audio sample represents sound measured during a sample time interval. The method further includes deriving an audio level from the audio sample and comparing the audio level to a threshold level. The audio level represents an average power level of the audio sample. The method further includes determining the talk/listen state depending on a relation of the audio level to the threshold level and depending on whether the audio sample includes voiced sound.
111 Citations
37 Claims
-
1. A method for determining a talk/listen state using voice detection, comprising:
-
receiving an audio sample representing sound measured during a sample time interval; detecting whether the audio sample includes voiced sound; deriving an audio level from the audio sample, the audio level representing an average power level of the audio sample; comparing the audio level to a threshold level; determining the talk/listen state depending on a relation of the audio level to the threshold level and depending on whether the audio sample includes voiced sound. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24)
-
-
25. An apparatus comprising:
-
a voice detection unit detecting whether an audio signal includes voiced sound responsive to receiving the audio signal; a talk/listen determination unit coupled to the voice detection unit, the talk/listen determination unit deriving an average audio power level of the audio signal and deriving a dynamic threshold level based on the average audio power level and past average audio power levels responsive to receiving the audio signal, the talk/listen determination unit determining a talk/listen state depending on a comparison of the average audio power level and the dynamic threshold level and on whether the voice detection unit detects voiced sound. - View Dependent Claims (26, 27, 28, 29, 30, 31, 32, 33, 34, 35)
-
-
36. A voice activated switching device for selecting a primary transmission site from among a plurality of transmission devices, the voice activated switching device comprising:
-
means for determining whether audio signals received from each of the transmission devices include voiced or unvoiced sound; means for repeatedly determining a dynamic threshold level for each of the transmission devices; means for comparing each of the audio signals received from each of the transmission devices to a corresponding dynamic threshold level; means for determining a talk/listen state for each transmission device based on whether each audio signal includes voiced sound and on whether a power level of each audio signal is greater than the dynamic threshold level. - View Dependent Claims (37)
-
Specification