next up previous contents index
Next: Training Up: Speaker verification/identification Previous: Speaker characteristics verification or

Error measure

Speaker verification errors

The different errors are summarised in Table 2.3.

 

  
ACCEPTED REJECTED
The speaker is The system accepts The system rejects him
user U him as speaker U (false rejection)
The speaker is an The system accepts The system rejects him
impostor  who claims him as speaker U
to be user U (false acceptance)
Table 2.3: False acceptance versus false rejection 

The application developer may have access to the different parameters that determine the system performance. These parameters may be directly correlated to given threshold or confidence levels. For example, he may accommodate the thresholds in order to obtain an equal error of false acceptance and false rejection  as well as a confidence area  as depicted in Figure 2.6. The technology provider should indicate how to handle such parameters. He also should indicate whether such parameters are speaker independent   or speaker dependent. If these are speaker-dependent parameters then he should know how to obtain and optimise them.

 figure1370
Figure 2.6: False acceptance  versus false rejection  

Speaker identification error

The speaker identification process depends on the size of the population, and the criteria indicated above for the speaker verification process should account for that. The error measures should also consider the confusion  that may occur between two different speakers (the substitution  possibility of identifying user E instead of the present user U). A particular summary is given in Table 2.4.

 

ACCEPTED SUBSTITUTED  REJECTED
The speaker is The system accepts The system accepts The system
the user U him as speaker U him as user E rejects him
(substitution) (false rejection )
The speaker is The system accepts The system accepts The system
an impostor  him as speaker U him as user E rejects him
(false acceptance ) (substitution)
Table 2.4: Different types of error 

Substitution  errors are more severe than the others because unauthorised speakers or impostors  may thus gain access to confidential data. As the prime motivation for integrating speaker identification procedures is to achieve more reliable personal identification in a convenient manner this has to be used with other techniques.

In practical application one imagines that the users are motivated and hence are very cooperative. Meanwhile the impostors  are unknown speakers and there is no way to collect data to prepare a rejection model   based on an ``impostor model''.  

The application developer has also to know how to calibrate the different thresholds to obtain the best compromise between false acceptance , false rejection  and substitution  errors.

In some particular applications dealing with speaker verification, a confusion matrix  may be requested in order to allow a pre-selection of several candidates in a first phase and then consider a second process with a small subset of speakers.



next up previous contents index
Next: Training Up: Speaker verification/identification Previous: Speaker characteristics verification or

EAGLES SWLG SoftEdition, May 1997. Get the book...