MESIE 16-Channel Spatial Audio Beamforming Engine (MESIE-MultiAudio-v1)

Studio 1.4M-Parameter PyTorch Transformer for 16-Channel Direction-of-Arrival (DoA) Tracking, Acoustic Beamforming, and Active Noise Cancellation.


Executive Overview

MESIE-MultiAudio-v1 is a spatial acoustic processing model powered by 1,437,584 real PyTorch parameters. Built for Acoustic Engineers and Spatial Audio Systems, it processes 16-channel microphone array signals to track 3D Direction-of-Arrival (DoA) azimuth/elevation angles, isolate target voices, and apply active noise cancellation (ANC).


Integration Example

from agent_helper import MESIEMultiAudiov1Agent

agent = MESIEMultiAudiov1Agent()

spatial = agent.perform_16channel_beamforming_and_doa(microphone_array_16chan=[[0.1]*16]*10)
print(spatial)

License

Apache 2.0

Downloads last month
85
Safetensors
Model size
1.44M params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support