Multi-modal synthetic voice detection system and method

A system, comprising:

• a communications interface configured to receive an audio input from a voice of a caller initiating a call;
• a processing system including a processor;
• and
• a memory that stores executable instructions that, when executed by the processing system, facilitate modules configured to perform operations, the modules comprising:
• a passive detection module configured to analyze the audio input to extract acoustic features, spectral patterns, prosodic elements, and lexical characteristics using passive techniques;
• an active engagement module comprising an active controller agent and at least one agent configured to initiate interactive challenges with the caller in real-time, wherein the interactive challenges including current event inquiries, logic puzzles, interruption handling tests, and requests for non-speech sounds;
• an automatic speech recognition module configured to transcribe responses from the caller;
• a natural language processing module configured to evaluate the responses for response timing, content coherence, complexity, and logical consistency;
• a composite score generator configured to combine outputs of the passive detection module and the active engagement module to derive a composite synthetic voice detection score;
• and
• a decision module configured to determine, based on the composite synthetic voice detection score, whether the voice of the caller is generated by artificial intelligence.

A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processing system including a processor, facilitate performance of operations, the operations comprising:

• receiving an audio input from a voice of a caller;
• analyzing the audio input to extract acoustic features, spectral patterns, prosodic elements, and lexical characteristics using passive detection techniques;
• actively engaging in a conversation with the caller, wherein the conversation presents an interactive challenge;
• receiving responses from the caller to the interactive challenge;
• processing the responses to assess response timing, content coherence, and complexity using automatic speech recognition and natural language processing techniques;
• combining the passive detection techniques with an evaluation of the responses of the caller to generate a composite synthetic voice detection score;
• and
• based on the composite synthetic voice detection score, determining whether the voice of the caller is generated by artificial intelligence.