Automatic gain control for adjusting gain during non-speech portions

1. A method comprising

estimating power of a speech portion of a speech signal that includes speech portions separated by non-speech portions, the power for the speech portion being estimated based on a power envelope that spans the speech portion, and

refraining from adjusting the gain of an automatic gain control during the speech portions.

2. The method of claim 1 in which each of the non-speech portions comprises silence.

3. The method of claim 1 in which the each of the speech portions comprises a word.

4. The method of claim 1 in which, for each of the speech portions, the gain is controlled to be constant.

5. The method of claim 1 in which the gain is controlled so as to be adjusted during non-speech portions.

6. The method of claim 1 in which the estimating includes estimating a power of the speech signal separately for each of the speech portions.

7. The method of claim 6 in which the estimating includes averaging the estimated powers of the speech portions.

8. The method of claim 1 in which the estimating includes detecting a maximum power that occurs during each of the speech portions.

9. The method of claim 1 also including detecting voice activity as an indication of the start of each portion of the signal.

10. A method comprising

detecting voice activity as an indication of the start of each portion of a speech signal, the speech signal including word portions separated by pause portions,

estimating power for the word portions, and

controlling the gain of an automatic gain control to be constant for the duration of the word portions and to be adjusted, based on the estimated power, during the pauses between the words.

11. A method comprising

estimating power of a speech portion of a speech signal that includes speech portions separated by non-speech portions, the power for the speech portion being estimated based on a power envelope that spans the speech portion, and

controlling an automatic gain control to be adjusted during the non-speech portions based on the power estimate.

12. The method of claim 11 in which estimating the power for each of the speech portions comprises estimating a peak power level.

13. The method of claim 11 in which the estimating includes averaging the power estimates for the speech portions of the signal.

14. The method of claim 11 also including detecting a presence or absence of voice activity as an indication of the boundaries of the speech and non-speech portions.

15. The method of claim 11 also including adjusting a gain of an AGC based on the estimating of the power of the speech signal.

16. Apparatus comprising

a port to receive a speech signal, and

an automatic gain control configured to apply a constant gain to a speech portion of the signal and to adjust the gain during non-speech portions of the signal based on power estimates done during previous speech portions.

17. The apparatus of claim 16 in which the automatic gain control includes power estimating elements configured to generate an estimate of a power of the speech portions of the speech signal.

18. The apparatus of claim 16 in which the automatic gain control includes voice activity detection elements.

19. A medium carrying instructions capable of causing a machine to estimate power of a speech portion of a speech signal that includes speech portions separated by non-speech portions, the power for the speech portion being estimated based on a power envelope that spans the speech portion, and

refrain from adjusting the gain of an automatic gain control during the speech portions.

20. The medium of claim 19 embodied as firmware associated with a digital signal processing chip.

21. A system comprising

a port to receive speech signals,

an automatic gain control configured to

estimate power of a speech portion of a speech signal that includes speech portions separated by non-speech portions, the power for the speech portion being estimated based on a power envelope that spans the speech portion, and

refrain from adjusting the gain of an automatic gain control during the speech portions, and elements configured to perform speech functions based on an output of the automatic gain control.

22. The system of claim 21 embodied in a multi-channel voice processing board.

23. Apparatus comprising

means for estimating power of a speech portion of a speech signal that includes speech portions separated by non-speech portions, the power for the speech portion being estimated based on a power envelope that spans the speech portion, and

means for refraining from adjusting the gain of an automatic gain control during the speech portions.