Apparatus and method for automatic gain control

摘要

A speech signal is received at an input. At least one electrical value associated with the received speech signal is tracked. A dynamic adjustment of the speech signal is determined. The dynamic adjustment is selected at least in part so as to minimize a distortion and minimize an over-amplification of the speech signal based at least in part upon an analysis of the at least one electrical value. The dynamic adjustment is further selected to obtain a desired output signal characteristic for the speech signal presented at an output. The dynamic adjustment value is applied to the speech signal and the adjusted speech signal is presented at the output. The gain of the signal can also be limited to prevent over-amplification.

主权项

1. A method for automatic gain control of speech signals, the method comprising:
receiving a speech signal at an input; computing a power level for a first speech frame that is obtained from the speech signal received at the input; determining a slow envelope tracking estimate of the computed power level for at least the first speech frame,
wherein determining the slow envelope tracking estimate of the computed power level for at least the first speech frame further comprises following a rising signal edge faster than following a falling signal edge; determining a fast envelope tracking estimate of the computed power level for at least the first speech frame,
wherein determining the fast envelope tracking estimate of the computed power level for at least the first speech frame further comprises detecting rising speech signal power levels more rapidly than detecting falling speech signal power levels; comparing the slow envelope tracking estimate to the fast envelope tracking estimate; determining a speech frame to be speech that is starting, when the fast envelope tracking estimate increases faster than the slow envelope tracking estimates and determining a speech frame to be speech that is ending, when the fast envelope tracking estimate decreases faster than the slow envelope tracking estimate; estimating a peak power level for speech in the first speech frame, the peak power level in a frame being determined responsive to a determination of speech being in the first speech frame adjusting a gain provided to a speech signal during a second speech frame responsive to determining an estimate of the speech peak power level for the first speech frame, the gain adjustment provided to the second speech frame being made while speech is present in the speech signal and approximately equal to a quotient of a desired output power level of the speech signal and the computed power level of the first speech frame; detecting a speech pause associated with an electrical envelope and not applying the gain adjustment while speech is not present thereby preventing amplifying noise more than amplifying speech when background noise is present and speech is not present; and presenting an amplified speech signal at the output; wherein said method is performed by a processor.