Matching voice samples to an entire database of voice recordings is a tedious job, which must be performed with full concentration—an easy way out is to outsource the listening to technology. With the help of GoVivace’s Speaker Identification solution users just simply provide the software with the voices to be matched and it returns the voices from the available recordings that come close to matching the target set. We provide you with the Software Developer Kit (SDK) library as well as the websocket with bidirectional streaming to use the software as a service. The speech to text engine is designed to work 24 X 7 X 365 and can be used to build mobile, web as well as high volume application.
The GoVivace’s Speaker Identification solution can work in both offline as well as online mode. The solution’s offline mode is better suited to forensic scientists who need to process large directories of voice samples. It can also be deployed in a live face-to-face situation where a covert security check is mandated on an individual. Then, it is easier to compare the person’s voice with available recordings rather than check other common biometric samples—the iris or the fingerprint—which must be expressly taken.
The solution’s online mode enables the rapid identification of callers. When a call comes in, the solution works rapidly in the background, comparing the caller’s voice with voice recordings of known fraudsters stored in a database, as the front-line agent conducts the usual telephonic check of the caller. Based on the outcome of this process and the agent’s own checking, the agent can decide how to process the caller’s request. GoVivace’s Voice Identification technology is language and channel independent, accepting samples of spoken speech as well as microphone samples.
GoVivace’s Voice Identification technology is available for Linux, Windows and Mac platforms and is available both as a desktop solution as well as a client/server application.
The GoVivace’s Voice Identification desktop solution comes with a graphical user interface that allows users to specify the sample voice or voices, the directory of voice recordings to process and other elements of the matching process. It also allows fresh voices to be appended, through file transfers, to the specifications of the running application—a useful feature for heavy duty intelligence operations.
The GoVivace’s Voice Identification client/server application comes with an Application Programming Interface offering similar functionality as the desktop version. Clients can connect to the server on demand.
The GoVivace’s Speaker Identification Solution is available in both 32 and 64-bit versions for Linux, Windows, and Mac platforms. A minimum of 4GB of RAM and a 2.0GHz processor is recommended.