Research Assistant (master thesis worker): universal speech enhancement for speech-based health biomarkers

Aalto University is where science and art meet technology and business. We shape a sustainable future by making research breakthroughs in and across our disciplines, sparking the game changers of tomorrow and creating novel solutions to major global challenges. Our community is made up of 120 nationalities, 14 000 students, 400 professors and close to 5000 faculty and staff working on our dynamic campus in Espoo, Greater Helsinki, Finland. Diversity is part of who we are, and we actively work to ensure our community’s diversity and inclusiveness. This is why we warmly encourage qualified candidates from all backgrounds to join our community.

The Department of Information and Communications Engineering at Aalto University carries out research and education in signals, machine learning, speech and audio technology, and human-centered information processing. In Professor Paavo Alku's research group, we develop methods and models for robust speech technologies, including applications in speech-based health assessment.

We are now looking for a paid Research Assistant as master's thesis worker in universal speech enhancement for speech-based health biomarkers.

Are you excited about deep learning for speech and want your thesis to contribute to future health technologies?

We are looking for a master’s thesis worker to study the role of universal speech enhancement in speech-based biomarking of health. Speech-based biomarking aims to infer health-related information from speech signals (e.g., voice quality, articulation, prosody) and has been investigated for a range of conditions and states such as neurological disorders (e.g., Parkinson’s disease, ALS), respiratory illnesses, mental health (e.g., depression, stress), and voice and speech disorders. This enables non-invasive and scalable monitoring, screening, and follow-up.

A key challenge is robustness: biomarker models are often trained on controlled recordings, but in real use the speech may be captured with different microphones and in varying acoustic environments. For example, a model trained on noiseless, close-talking recordings can fail when used with a smartphone in a reverberant kitchen or a noisy café due to a mismatch between training and deployment conditions.

Universal speech enhancement refers to methods designed to improve speech across a broad range of real-world distortions (noises, reverb, limited bandwidth, codecs, clipping, etc.) without being tailored to one specific environment or distortion type. While universal enhancement has been widely explored for speech communication and automatic speech recognition, its impact on speech-based health biomarkers has not been systematically studied. This thesis addresses that research gap. In this project, you will examine whether such enhancement can act as a robust front end that improves the consistency of the speech signal before biomarker feature extraction, thereby reducing the impact of recording-condition mismatches.

Your role and goals

In this position, you will design and run experiments that combine universal speech enhancement with downstream biomarker tasks. The work is research-oriented and hands-on, and will contribute to methods developed in the group. Typical tasks include:

• Document the work and write the master's thesis. Results may also be prepared for publication.

• Evaluate performance of solutions in speech-based biomarking tasks and analyse results.

• Run simulations and enhancement experiments covering realistic environmental variability (noise, reverberation, speech codecs, etc.).

• Adapt and benchmark publicly available enhancement baselines.

• Contribute to existing methods developed in the research group.

Your network and team

You will work in Professor Paavo Alku's research group at Aalto University. Your thesis supervisor will be Professor Paavo Alku, and your day-to-day advisor will be Saska Tirronen. You will also collaborate with other researchers in the group and benefit from an active research environment.

Your experience and ambitions

We are looking for a candidate who is motivated to learn and to build reliable experimental pipelines. The ideal candidate has:

• Final-year M.Sc. student in information technology, electrical engineering, or a related field (speech/signal processing).

• Good understanding of deep learning and prior practical experience with neural networks.

• Basics of speech processing (e.g., ELEC-E5500 Speech Processing or equivalent).

• Proficiency in PyTorch.

• Proficiency in Python programming.

Language requirements:
English: Working proficiency in English is required.
Finnish: Finnish language is not required. Finnish proficiency is considered an advantage.

If you are looking for a thesis at the intersection of speech enhancement, machine learning, and digital health, this position offers a clear research question with real-world relevance and strong academic supervision.

What we offer
We offer a supportive environment for completing a high-quality thesis and building practical research skills. In this role, you will benefit from:

• Paid employment for master's thesis work (salary: 2565 EUR/month).

• Close supervision and support from the research group.

• A timely research topic and a clear research gap at the intersection of universal enhancement and speech-based health biomarkers.

• Hands-on work with modern deep learning methods, reproducible experimentation, and rigorous evaluation.

• An international research community at Aalto University in Otaniemi, Espoo.

Start date is as soon as possible. Working arrangements are flexible and will be agreed based on your thesis schedule.

Join us!
Please submit your application as a single PDF file by 4.2.2026 through our recruitment system using the “Apply now!” link below.

Your application should include:

• Letter of motivation (max. 1 page)

• Curriculum Vitae (CV)

• Transcript of grades

Please note that the position is open only to Aalto University M.Sc. degree students.

We will review applications on a rolling basis and will hire the right person as soon as we find them, so please apply early.

Aalto University’s employees should apply for the position via our internal HR system Workday (Internal Jobs) by using their existing Workday user account (not via the external webpage for open positions). If you are a student or visitor at Aalto University, please apply with your personal email address (not aalto.fi) via Aalto University open positions

For more information about the role, please contact Doctoral Researcher Saska Tirronen (advisor) by email: saska.tirronen@aalto.fi. In recruitment process related questions, please contact HR Advisor Johanna Haapalainen (hr-elec@aalto.fi).

More about Aalto University:

Aalto.fi
youtube.com/user/aaltouniversity
linkedin.com/school/aalto-university/
www.facebook.com/aaltouniversity
instagram.com/aaltouniversity

To view information about Workday Accessibility, please click here.

Please see more of our Open Positions here.