Verbal command processing based on speaker recognition

1. An apparatus comprising:

a processor;

a memory that stores code executable by the processor to:

select recognition phonemes from a phoneme input stream that correspond to stored recognition phonemes by segmenting the phoneme input stream using an average level crossing rate to determine each recognition phoneme;

select a first user profile from a plurality of user profiles for a first user recognized based on the recognition phonemes, wherein each user profile comprises one or more permissions, one or more customized actions, and one or more speech profiles, each permission allows the execution of an associated command, and each customized action is only executed if the first user profile is active; and

process a command concurrently identified from the phoneme input stream from a second user in response to the first user profile comprising a permission for the first user to execute the associated command;

retrieve a customized action associated with the command from the first user profile; and

in response to the first user profile being active, execute the customized action.

2. The apparatus of claim 1 wherein the code is further executable by the processor to recognize the first user and the second user from a comparison of the recognition phonemes to the stored recognition phonemes.

3. A method comprising:

selecting, by use of a processor, recognition phonemes from a phoneme input stream that correspond to stored recognition phonemes by segmenting the phoneme input stream using an average level crossing rate to determine each recognition phoneme;

selecting a first user profile from a plurality of user profiles for a first user recognized based on the recognition phonemes, wherein each user profile comprises one or more permissions, one or more customized actions, and one or more speech profiles, each permission allows the execution of an associated command, and each customized action is only executed if the first user profile is active;

processing a command concurrently identified from the phoneme input stream from a second user in response to the first user profile comprising a permission for the first user to execute the associated command;

retrieve a customized action associated with the command from the first user profile; and

in response to the first user profile being active, executing the customized action.

4. The method of claim 3 , the method further comprising:

segmenting an audio signal into the phoneme input stream;

recognizing the first user and the second user from the recognition phonemes; and

identifying the command from the phoneme input stream.

5. The method of claim 3 , the method further comprising normalizing vocal track lengths of phonemes of the phoneme input stream.

6. A program product comprising a non-transitory computer readable storage medium that stores code executable by a processor to perform:

selecting recognition phonemes from a phoneme input stream that correspond to stored recognition phonemes by segmenting the phoneme input stream using an average level crossing rate to determine each recognition phoneme;

selecting a first user profile from a plurality of user profiles for a first user recognized based on the recognition phonemes, wherein each user profile comprises one or more permissions, one or more customized actions, and one or more speech profiles, each permission allows the execution of an associated command, and each customized action is only executed if the first user profile is active;

processing a command concurrently identified from the phoneme input stream from a second user in response to the first user profile comprising a permission for the first user to execute the associated command;

retrieve a customized action associated with the command from the first user profile; and

in response to the first user profile being active, executes the customized action.

7. The program product of claim 6 , the processor further:

segmenting an audio signal into the phoneme input stream;

recognizing the first user and the second user from the recognition phonemes; and

identifying the command from the phoneme input stream.

8. The program product of claim 6 , the processor further normalizing vocal track lengths of phonemes of the phoneme input stream.