IP PBX – BPO Call center – Artatel
IP PBX – BPO Call center – Artatel
What is Voice Biometrics
Voice biometric technology that is able to verify the speaker's identity as a substitute for a PIN or finger print. Otherwise known as voice recognition or voice authentication. Currently this technology is called Voice Biometrics or voice biometrics. (passive voice biometrics, google voice biometrics)
The main advantage of voice biometrics is:
• Easy and fast access to information
Time consuming to ask customers for personal data by agents no longer occurs, so time
service per customer is much faster.
• Much safer
Your customer's confidentiality is guaranteed safe even by your own call center agent.
• Much faster, Voice biometrics takes less than 7 seconds to verify
• Can be used in various media
voice biometric authentication java voice biometric system voice biometrics in voice identification
Other advantages:
a. Time reduction authentication
Previously, every time your CS wanted to ensure that the caller was the legitimate account owner, he would
It takes a long time to carry out a tiring question and answer process such as:
– Ask for full name according to KTP
– Asking the biological mother's maiden name
– 4 digit credit card number
- Home address
- Place and date of birth
- Mobile phone number
- Email address
– Are there any additional cards
b. Improved customer experience & confidentiality.
Customers only need to state their full name, the server will determine whether the customer's voice is the owner
legitimate.
c. Money savings due to the added level of fraud prevention.
d. Lower operational costs.
The shorter authentication time means there is no need for the number of agents, workplaces and
other means more
e. Higher security and lower costs
f. Customers are happy and employees are satisfied because they don't experience personal questions every time they contact customer care.
g. A less secure and accurate alternative that requires additional hardware costs and physical presence
To start, it's important to know that no biometric is 100% accurate. Example a 2014 study on iris recognition determining system accuracy can be between 90 and 99%, a wide range. Voice biometric accuracy also falls within this range for various reasons. However, even with its imperfections, voice biometrics are an invaluable tool.
The method for assessing the accuracy of a voice biometric system is “Equal Error Rate”, or “EER”. EER is the point at which the “False Acceptance Rate” or “FAR” (i.e., letting fraudsters through) equals the “False Rejection Rate” or “FRR” (i.e., denying access to valid users). Voice biometric scoring systems are based on statistical probability, so there is a trade-off between these errors that needs to be considered. For example, if you set the trust level to “high” to prevent fraudsters, you may end up blocking more valid users, causing disruption. Setting a “lower” trust level will provide more comfort for your valid users, but you may let in more fraudsters. VBG will work with you to provide the optimal balance between security and comfort. Note also that EER is measured for a single trial. By allowing retries in your application, you can increase the likelihood that a valid user can get past a second or third retry, even if you initially set a high level of confidence to reject fraudsters.
Note that the EER, FAR, and FRR results come from the set of audio samples used to process and obtain these measurements. Beware of very low EER values advertised by some vendors, as laboratory-derived EER results can be easily manipulated by removing samples that negatively impact the results. EER results are only as good as the data sampling performed for their calculations. Unexpected real-world results can (and will) occur if your sampling is not truly representative of the end-user population, the particular language and dialect, the type of device used, the environment in which speech was collected, etc. Therefore, we recommend that you compare voice biometric systems based on real-world users, including running trials in your desired production environment,
Other important factors influencing real-world EER are the content of speech samples and the presence of noise. If the speech sample is too noisy, or if incorrect information is spoken to the voice biometric system, then the voice biometric engine will have difficulty using the speech sample to make accurate determinations. Using different types of devices can also affect the results.
For example, mobile phone networks use different compression techniques compared to landlines – this affects the voice biometric process that extracts unique vocal characteristics from speech samples.