Industry: Audio Region: Spain Transaction price: US$ 1 million Transaction method: equity investment
Project introduction:
Founded in 2005, this Spanish small and medium-sized enterprise has strong research and development capabilities in voice and audio technology, and provides customized solutions for customers in voice biometrics, artificial intelligence voice clone detection, voice recognition and audio watermarking.
● Voice biometrics (speaker verification and recognition):
The voice biometric technology developed by the company is an advanced speaker verification technology. Thanks to the company's text-independent speech biometric engine. Speaker verification (1:1 matching for authentication applications) and speaker recognition (1:N matching for comparing unknown voice with all available voiceprints) are supported. MS (matching score) and mass score for voiceprints are also provided. Once the voiceprint is created, even in different languages, it only takes 3 seconds to speak freely to identify the speaker.
● AI voice clone detection:
Due to the new progress of complex artificial intelligence deep forgery voice cloning technology, fake news, bank robbery and phishing attacks using voice simulation are increasing. Modern and advanced machine learning makes artificial voice almost indistinguishable from simulated human voice. The company's innovative AI voice clone detection technology can detect artificial sounds used by cyber criminals and scammers in real time. The different technologies of speech synthesis and cloning are modeled and detected. Even for the new advanced vocoder based on artificial intelligence. This technology can also be combined with the company's voice biometric technology as an advanced anti-fraud technology in the application of voice biometric authentication.
● Audio watermarking: Inserting inaudible information into the audio signal.
These messages are inaudible and cannot be extracted by unauthorized listeners, and they have strong anti-interference ability against channel distortion and compression (such as MP3 or recoding). It can even withstand digital-to-analog conversion. In addition, the watermark can also detect operations such as cutting and pasting audio clips.
● These technologies have a wide range of applications, including:
-Forensic acoustics of law enforcement agencies.
-Secure authentication in call centers and mobile/web applications.
-Car infotainment.
-Physical access control.
-Streaming media content protection.
-Advanced artificial intelligence vocoder detects simulated attacks on loudspeakers.
-Developing non-invasive remote measuring tools for intelligent manufacturing.
Cooperation mode:
The company provides customized services and technical authorization cooperation. Welcome to discuss.