The challenge

The challenge of automating lyric generation and ensuring accurate coverage for all songs is a critical problem in the music industry. With the vast amount of music being produced, including AI-generated tracks, and the variations in recordings (e.g., live, radio edits, remixes), finding precise and matching lyrics for audio recordings is increasingly complex. This issue is compounded by the need for accuracy in applications such as explicit content detection, topic classification, semantic search, and personalized recommendations. While advancements in Automatic Speech Recognition (ASR) offer promising techniques, applying them to music introduces unique challenges, such as handling noise, hallucinations, and other model limitations. Addressing these challenges is essential for improving the accessibility and utility of lyrics in the evolving music landscape.

The thesis

As a master thesis student your task is to contribute to the field of Automatic Lyric Transcription (ALT), presumably through building on state-of-the-art methods within Automatic Speech Recognition (ASR) and adapting to the specific challenges of singing and music audio.

Out-of-the-box state-of-the-art models for ASR, like Whisper (OpenAI) and Canary (Nvidia), transfer surprisingly well to the field of ALT. However, the existing challenges with hallucinations connected to silent and noisy audio segments are worsened in the context of music. Tackling these hallucinations through a novel method is a definite path to achieving robust and precise automatic lyric transcription.

As part of the music experience team, you will have the opportunity to work closely with other machine learning engineers and music experts on the task.

About you

You're driven and entrepreneurial, but you know how to be a team player too. Regardless of roles, we're always looking to work with people who can adapt to constant change, prioritize what's important, stay humble, open, curious, and have a passion for details.

This internship opportunity is for you if the following describes you:

• Passionate about data science and machine learning
• Able to communicate complex topics in a simple way
• Experience in Python - either academically, personally, or professionally
• Fluent in English
• Has residency or citizenship in Sweden, preferably near Stockholm

About us

Soundtrack Your Brand is a B2B scale-up company providing music streaming services for businesses. We serve small customers like the café around the corner, and much bigger brands like McDonald's, Toni & Guy, and TAG Heuer. On the inside, we're a bunch of talented, motivated, and humble designers, engineers, and music experts. We believe in product-led growth, where the product is the primary driver of customer acquisition, conversion, and expansion.

The team

The Music Experience team is dedicated to the ownership and development of all music features, such as music onboarding, music discovery (home, search, browse, and detail views), user-created playlists, and schedules. The team owns the features from UX/UI to APIs, databases, and machine learning. You can download our app and sign up for a free trial to get a first-hand experience of our features.

The position

A Master's Thesis is an excellent way for us to get to know new talent. We believe that diversity of perspective and experience makes our team and our product better, and we encourage you to apply.

This is a full-time thesis project intended for the Spring 2026 semester. If this sounds like the perfect final project of your master's degree, and a challenge you'd love to take on, we encourage you to apply. We reserve the right to close this vacancy early if we identify a suitable candidate before the application deadline. To ensure consideration, we encourage you to submit your application as soon as possible.

If you have any questions about the position or need to reach out, get in touch with Anton Cakste at anton@soundtrack.io. Please note that we only accept applications submitted via our career page and do not accept applications by e-mail.