We are excited to announce that EaseAudio has successfully deployed our machine learning–based noise suppression algorithm on the Cadence Neural Network Engine (NNE) integrated in the Airoha AB1595 Bluetooth audio platform.
This achievement demonstrates our capability  to deliver production-ready AI audio solutions optimized directly for silicon.

✅Powering Next-Generation TWS with AI

Airoha AB1595 is a leading Bluetooth audio SoC used in premium TWS earbuds, wireless headphones, and smart wearable devices.

By leveraging the on-chip Cadence Neural Network Engine, EaseAudio enables:
- Clearer voice calls in noisy environments
- Lower power consumption
- Real-time AI inference
- No need for additional processors

AI noise suppression is no longer limited to high-end application processors — it now runs efficiently on compact Bluetooth SoCs.

✅From Model to Silicon — Production Ready

Many companies can design AI models. Few can deploy them efficiently on embedded hardware.

EaseAudio provides full-stack expertise:
- Model quantization and optimization
- INT8/INT16 inference tuning
- DSP + NNE hybrid workload distribution
- Real-time pipeline integration
- Memory and performance optimization

Results:
- Real-time 10 ms processing
- Significant DSP MIPS reduction
- Stable Bluetooth voice performance
- Production-level firmware integration

✅Why This Matters for OEMs

Better voice quality means better user experience.

With EaseAudio’s ML-based noise suppression on AB1595:
- Users experience clearer calls in cafés, streets, and offices
- Battery life remains optimized
- No BOM increase from external AI processors
- Faster time-to-market with a ready solution

We help OEMs move from concept to shipping product — faster.

✅A Strategic Capability for AI Audio

This milestone reinforces EaseAudio’s position as:

- A trusted embedded AI audio partner
- Experts in Cadence NNE acceleration
- Specialists in Bluetooth audio SoC integration
- Deliverers of production-ready AI firmware

We don’t just build algorithms. We make them run efficiently inside real consumer products.

✅Looking Ahead

EaseAudio continues to expand our embedded AI capabilities:

- Multi-mic AI noise suppression
- AI beamforming
- Low-power always-on pipelines
- Cross-platform deployment across major SoCs

As AI becomes standard in audio devices, optimization at the silicon level becomes critical.

EaseAudio is ready.