Products are systems integrated into a specific application. Examples are given here to emphasise the benefits for both users and service providers.
Some telephone operators provide a toll free service allowing customers to place a phone call from any terminal. The charge for the call appears later on the phone bill of the user. For such an application, the user must be identified. A personal identification number is usually requested. This information could be entered via the dial pad, but the security of such a protocol is very low (a PIN could be easily stolen).
The SPRINT operator in the USA proposes a successful alternative. It uses speaker verification . Upon a call to the service, the user is requested to say the sequence of digits of his account. Digit recognition and speaker verification is performed on this sequence. In case of doubt, he is then prompted a random sequence of digits for further validation (getting some assurance that he is not playing back some prerecorded speech). He is then allowed to dial the phone number of the person he wants to reach.
The SPRINT operator has registered 1.5 million customers for this service. In such an application, a rather high false acceptance rate is tolerable. Impostors are aware that their voice is being recorded. The introduction of speaker verification proved to be dissuasive against fraud. With a rather small increase in the complexity of the access protocol, a satisfactory level of security was achieved. Of course, a higher degree of security could be reached by letting the user dial by voice. The risk of recognition error (although rather small as recognition could be performed in a speaker dependent mode) would necessitate confirmation and a longer dialogue (paid by the service provider!).
In the UK, the BT home banking trial with the Royal Bank of Scotland used speaker verification technology to provide an additional layer of security to PINs (Personal Identification Number). The concern of offending as few genuine customers as possible was of prime importance to the bank.
These examples show that the benefit of a speaker verification system is obtained as a compromise between the reduction of fraud and user acceptability. In particular, it is clear that the equal error rate usually does not correspond to a realistic functioning condition.