CS BSc Thesis Presentation 6 May 2026

Wednesday, 6 May there will be a bachelor thesis presentation in Computer Science at Lund University, Faculty of Engineering.

The presentation will take place in E:2405 (Glasburen).

Note to potential opponents: Register as an opponent to the presentation of your choice by sending an email to the examiner for that presentation (firstname [dot] lastname [at] cs [dot] lth [dot] se). Do not forget to specify the presentation you register for! Note that the number of opponents may be limited (often to two), so you might be forced to choose another presentation if you register too late. Registrations are individual, just as the oppositions are! More instructions for opponents are found here on the LTH thesis project page.

13:15-14:00 in E:2405 (Glasburen)

• Presenters: Ross Ottosson, Ludvig Ölund Danielsson
• Title: Automatiserad tabulaturgenerering: Förbättrad träffsäkerhet genom domänspecifik filtrering och Viterbi-optimering
• Examiner: Mattias Nordahl
• Supervisor: Ester Daniel Ytterbrink (LTH)

Automatic Music Transcription (AMT) of the guitar is a significant challenge due to the instrument's acoustic complexity, including overtones, sympathetic resonance, and mechanical transients. This study investigates how a general-purpose AMT model (Spotify’s Basic Pitch) can be adapted to generate physically playable guitar tablature. By lowering the model's minimum note length threshold to 40 ms, fast musical passages are successfully captured, but this results in a "shotgun effect" of massive over-detection. 

To mitigate this, a domain-specific filtering pipeline was developed and optimized based on the physical and acoustic properties of the guitar. The purified MIDI data was subsequently processed using a Viterbi algorithm to calculate biomechanically ergonomic fingerings. Results indicate that the customized pipeline outperforms standard Basic Pitch by increasing precision and F1 scores. Furthermore, the filtering significantly reduced the Viterbi algorithm’s crash rate from 11.13% to under 1% by eliminating physically impossible note combinations.