Thesis - Intelligent Automatic Audio Adjustment

Background noise can interfere with communication, while conventional automatic volume control reacts only once noise levels have changed. This thesis at Axis Communications investigates whether video-based scene information can predict such changes and enable earlier audio adjustments. The aim is to achieve smooth volume adaptation while preserving speech intelligibility during calls.

The assignment

• Investigate methods that use scene information, such as the number of people or objects present, to predict changes in background noise.
• Explore how these predictions can support proactive audio adjustment without compromising speech intelligibility.
• Potentially investigate whether combining information from nearby devices or different sensors within a device improves noise predictions.

Who we are looking for

• Studying towards a master's or bachelor's degree in Computer Science, Software Engineering or a related subject.
• Affiliated with a Swedish university or university college, either directly or through an exchange programme.

What we offer

• An open, friendly and collaborative working culture at a company developing network technology, audio systems and intelligent analytics.

Practical details

Location: Lund. Open to one or two master's students, or two bachelor's students. Apply through the thesis advert by 13 December 2026, in Swedish or English, with a CV and academic transcript. If applying as a pair, submit separate applications and name your intended thesis partner. Questions: Anders Olsson, +46-705-191441, Anders.Olsson@axis.com.