It is the ability to distinguish the voice of one individual from that of another. Voiceme
uses the most advanced deep learning technologies and proprietary models of voice
identification to:
- Extract the unique vocal characteristics of an individual (such as fundamental frequency, cadence, timbre, etc.)
- Compare the extracted vocal characteristics with a voice database: the database must include voice samples of any individuals the system needs to be able to identify.
- Select the database voice that most closely resembles the extracted vocal characteristics: the selected voice is therefore identified as the speaker’s voice.
Facial recognition is a biometric technology that uses artificial intelligence to identify an individual or verify a person's identity by analysing their facial features.
It includes two main steps:
- Face detection: the system detects one or more faces in a digital image or video.
- Facial recognition: the system compares the extracted facial features with those in Voiceme. If a facial feature match occurs, the person's identity is verified.
Voiceme generates a random sentence the user is asked to repeat during authentication process. Thanks to proprietary Speech-To-Text algorithms, the platform verifies the perfect matching of user words with the sentence to be repeated. Through this functionality, Voiceme is able to prevent fraud resulting from Synthetic Identities created with Generative Artificial Intelligence.
It is a method used in Voiceme biometric recognition systems to determine whether a biometric capture or sample comes from a live human being or an inanimate or unauthenticated source. This method enables recognition of a live person from a fraudulent presentation (Synthetic Identity) or a static image, thus minimizing the risk of fraud or fakeness.
Used in addition to random generation of sentences, this method makes the random sentence valid for a few seconds only as if it were a Voice OTP (One Time Password). A method that enhances biometric authentication security.
