Thesis – Speech-to-Text on Edge IoT Device

Growing demand for speech-to-text and advances in AI create opportunities to improve transcription accuracy and efficiency. This thesis at Axis Communications explores AI-supported speech-to-text on an edge IoT device and the architecture of a live-translation pipeline spanning edge and cloud.

The assignment

• Build an initial speech-to-text prototype for an edge device using existing tools and available AI models.
• Study the architecture of a live-translation pipeline from the edge device to the cloud.
• Compare approaches while accounting for hardware performance constraints.
• Make the solution readily accessible using modern version control or other suitable tools.

Who we are looking for

• You are studying Computer Science, Software Engineering or a related subject at bachelor's or master's level.
• You must be affiliated with a Swedish university or university college, either directly or through an exchange programme.

What we offer

• Work in a friendly, open and collaborative company culture that encourages new ideas and continued development of skills and working methods.

Practical details

Location: Lund. The project is intended for one or two master's students, or two bachelor's students. Apply through the thesis advert in Swedish or English, including your CV and university grade summary. If applying as a pair, submit separate applications and clearly identify your project partner. Application deadline: 13 December 2026. Questions: Anders Olsson, +46-705-191441 or Anders.Olsson@axis.com.