Speech recognition with sequence parsing, rejection and pause detection options

A recognition system comprising:

• input means for receiving a speech signal;
• recognition processing means for processing the speech signal to indicate its similarity to predetermined patters to be recognised;
• output means for supplying a recognition signal indicating recognition of one of said patterns;
• and
• rejection means for rejecting the recognition signal under predetermined conditions, the rejection means being arranged to receive at least one signal parameter derived from said speech signal which does not depend upon the output of said recognition means, wherein the recognition means is arranged to partition the speech signal into a pattern-containing portion and, preceding the following said pattern-containing portions, noise or silence portions, and the rejection means is further arranged to be responsive to said partitioning.

In a recognition system operating on a speech signal, an energy averager comprising:

• means for storing an energy level relating to previous energy levels of the speech signal;
• means for comparing the difference between the speech signal energy and said energy level with a threshold;
• means for varying the stored average energy level in response to the difference exceeding the threshold;
• and
• means for varying the threshold depending upon the difference.

A method of operating a speech recognition system which comprises input means for receiving a speech signal, recognition processing means for processing the speech signal and output means for indicating a recognized speech pattern, the method comprising the steps of:

• receiving a speech signal;
• processing the speech signal to generate a pattern signal identifying a predetermined pattern to which the speech signal is recognized as corresponding to partitioning the speech signal into a sequence of successive temporal portions and comparing said sequence of successive temporal portions with a predetermined sequence of pre-speech noise or silence portions;
• speech pattern portions;
• and post-speech noise or silence portions to generate said pattern signal;
• detecting the arrival of a point in time within said post-speech portion and after the beginning thereof, said detecting being in response to the identification of the onset of the post-speech portions in the recognition processing step and including repeatedly performing a detection operation which depends both on said onset and a signal parameter derived from said speech signal, which parameter is independent of said onset, and
• routing said pattern signal to said output means on detection of said point in time, to enable the immediate operation of utilizing apparatus connected thereto.

A method of operating a speech recognition system which comprises input means for receiving a speech signal, recognition processing means for processing the speech signal and output means for indicating a recognized speech pattern, the method being to detect the arrival of a point in time after the end of the speech pattern, the method comprising the steps of:

• pre-processing a temporal portion of a speech signal received at the input means;
• performing a recognition process on that temporal portion and preceding temporal portions to generate a pattern signal identifying a predetermined pattern to which the speech signal is recognized as corresponding;
• and
• recognizing whether the point in time has occurred;
• wherein the step of recognizing whether the point in time has occurred comprises:
• deriving at least one signal parameter from said speech signal which is independent of the partitioning between speech and noise performed by the recognition process;
• deriving at least one parameter which depends upon said partitioning;
• and deciding whether or not said point in time has arrived taking into account both parameters.