Analysing speech signals
A method of analysis of an audio signal comprises: receiving an audio signal representing speech; extracting first and second components of the audio signal representing first and second acoustic classes of the speech respectively; analysing the first and second components of the audio signal with models of the first and second acoustic classes of the speech of an enrolled user. Based on the analysing, information is obtained information about at least one of a channel and noise affecting the audio signal.
1. A method of analysis of an audio signal, the method comprising:
receiving an audio signal representing speech;
extracting first and second components of the audio signal representing voiced and unvoiced speech respectively;
analysing the extracted first component of the audio signal with a first model of voiced speech of an enrolled user and the second component of the audio signal with a second model of unvoiced speech of the enrolled user;
based on said analysing, determining at least one of a property of a channel affecting the first and second components of said audio signal and a property of noise affecting the first and second components of said audio signal; and
compensating the received audio signal for the property of the channel and/or the property of noise.
2. A method according to claim 1 , wherein extracting first and second components of the audio signal comprises:
identifying periods when the audio signal contains voiced speech; and
identifying remaining periods of speech as containing unvoiced speech.
3. A method according to claim 1 , wherein analysing the first and second components of the audio signal with the first and second models of the speech of the enrolled user comprises:
comparing magnitudes of the audio signal at a number of predetermined frequencies with magnitudes in the first and second models.
4. A method according to claim 1 , comprising:
performing a speaker identification process on the received audio signal to form a provisional decision on an identity of a speaker;
selecting the first and second models of the speech of the enrolled user, from a plurality of models, based on the provisional decision on the identity of the speaker;
compensating the received audio signal for channel and/or noise; and
performing a second speaker identification process on the compensated received audio signal to form a final on the identity of the speaker.
5. A method according to claim 4 , wherein compensating the received audio signal for channel and/or noise comprises:
identifying at least one part of a frequency spectrum of the received audio signal where a noise level exceeds a threshold level; and
ignoring the identified part of the frequency spectrum of the received audio signal when performing the second speaker identification process.
6. A system for analysis of an audio signal, the system comprising:
an input for receiving an audio signal; and
one or more processors configured to:
receive an audio signal representing speech;
extract first and second components of the audio signal representing voiced and unvoiced speech respectively;
analyse the extracted first component of the audio signal with a first model of voiced speech of an enrolled user and the second component of the audio signal with a second model of unvoiced speech of the enrolled user;
based on said analysis, determine at least one of a property of a channel affecting the first and second components of said audio signal and a property of noise affecting the first and second components of said audio signal; and
compensate the received audio signal for the property of the channel and/or the property of noise.
7. A non-transitory computer readable storage medium having computer-executable instructions stored thereon that, when executed by processor circuitry, cause the processor circuitry to perform a method according to claim 1 .