Home / Case Studies / Automotive
Case Study · Automotive

ML noise suppression for in-car voice control

A global automotive manufacturer needed in-car voice commands to stay accurate over engine and road noise. We built an ML-driven adaptive filtering system, now in final pre-production testing for a new vehicle line.

+40%Recognition accuracy in noise
95%Positive tester feedback
Real-timeOn-device inference
Pre-prodFinal testing

The challenge

Voice control degraded badly under real driving conditions — engine, road and wind noise pushed recognition error rates up and frustrated drivers. Fixed filters could not adapt to changing acoustic conditions.

What we built

Adaptive filter correction

Machine-learning-driven correction of IIR/FIR filter coefficients that adapts to changing cabin noise in real time.

Noise-robust training

Models trained across engine, road and wind-noise profiles representative of real driving.

On-device inference

Low-latency inference within automotive compute constraints, suitable for embedded deployment.

Validation harness

An evaluation harness measuring recognition accuracy across controlled noise scenarios for pre-production sign-off.

Results

Figures reflect outcomes measured on this engagement. Client withheld under NDA.

+40%recognition accuracy in noise
95%positive tester feedback

We treated this as a signal-processing problem solved with ML rather than a generic model — adapting proven filtering to changing acoustics. The system is in final pre-production testing ahead of a new vehicle line.

Related resources

ServiceAI & Machine LearningSignal processing and audio ML in productionServiceDevOps & PlatformMLOps and reliable deliveryCase StudyAI Dubbing10× faster localization

Have a similar project in mind?

We scope a clear plan with milestones and architecture options — and right-sized GPU hardware if AI workloads are involved.

sales@haink.org