AskJADA V2 Beta: AI Phone Agents Deliver Human-Level Responses with 83% Faster Processing Times
MomentPort Inc. is thrilled to announce that AskJADA V2 Beta has achieved a major milestone with an 83% decrease in average voice call processing time, dropping from 2700 milliseconds to 446 milliseconds. This improvement was reached through simultaneous multi-processing, enabling AskJADA to query multiple sources and execute multiple functions at once and return only the most logical result. Additionally, the voice streaming architecture was redesigned to send audio in smaller chunks, reducing network latency and accelerating delivery. These optimizations, along with other minor enhancements, led to this substantial reduction in processing times.

Another key enhancement in V2 Beta involves JADA’s memory architecture. Previously, JADA evaluated the relevance of each fact for both immediate responses and long-term storage in a strictly sequential manner, leading to a delay after each question-and-answer exchange. In V2 Beta, this process has been redesigned to function more akin to the human brain, allowing JADA to assess each fact simultaneously for its utility in generating a response and its potential value in long-term memory. This modification not only reduces latency, but also more closely mirrors the subconscious processes by which humans filter and retain information.
In addition to these major updates, we also optimized our AI servers used for JADA’s memory storage and reduced network latency further. While these improvements contribute significantly to better performance, accuracy has likewise seen notable gains through the implementation of simultaneous multi-processing. By enabling JADA to process a broader range of data when making decisions, the system now delivers even greater accuracy overall.
Audio Authenticity: The video above includes audio recorded via a Countryman lavalier mic and combined with a server-side recording for clarity. No timing or content was altered. The raw lav mic track is available here: raw_audio_v2_live_demo.mp3.
Meet JADA. Your business's first AI Representative.