Hi everyone,
I was wondering if anyone knows whether future versions of 3CX might include AI Agents or an AI Receptionist that would no longer rely on OpenAI, but instead take advantage of the hardware GPU already used for speech transcription.
With the recent improvements in on‑device speech processing, it seems technically possible to run lightweight AI models locally. This could offer more privacy, lower latency, and less dependence on third‑party APIs.
Has 3CX mentioned anything about exploring this direction, or is it something currently being considered for upcoming releases?
Thanks in advance for any insights!
I was wondering if anyone knows whether future versions of 3CX might include AI Agents or an AI Receptionist that would no longer rely on OpenAI, but instead take advantage of the hardware GPU already used for speech transcription.
With the recent improvements in on‑device speech processing, it seems technically possible to run lightweight AI models locally. This could offer more privacy, lower latency, and less dependence on third‑party APIs.
Has 3CX mentioned anything about exploring this direction, or is it something currently being considered for upcoming releases?
Thanks in advance for any insights!
