MSc in Voice Technology

This one-year master's in Voice Technology will give you the expertise you need to contribute to the cutting edge in voice synthesis and speech recognition.

This interdisciplinary program is suitable for students coming from the social sciences, computer sciences, and the humanities. The main requirement is interest, enthusiasm, and willingness to learn new things. Speech technologies have permeated life so much that we hardly even notice their presence. Whenever you dictate a message on your smartphone, ask Alexa the weather, or learn a language using an app, you are using voice technologies. And with new Internet-of-Things applications on the horizon, smart spaces and the presence of Siri, Alexa, Cortana, and Google Home are poised to grow even further.

This program, developed in consultation with partners from Silicon Valley, local start-ups, and universities worldwide, will ensure you acquire expertise in voice synthesis and speech recognition and have the requisite mastery of programming, machine learning, and speech sounds. You will work in teams to make your own synthetic voices, speech recognizers, and more. You’ll even make your own voice tech demo as a final project. People with scientific expertise in the domain of voice technology are in short supply. We aim to fix that!

Why study this program in Leeuwarden?

The Voice Technology MSc. is the only graduate program dedicated to voice synthesis and speech recognition. This is an interdisciplinary program in which students from diverse backgrounds collaborate with each other and industrial partners. You will develop speech recognizers, make synthetic voices, and execute empirical experiments. Your thesis project also includes making a demonstrator showcase your expertise.

Leeuwarden is a city of innovation, home to many start-ups and established companies, including serious game and tech businesses which could collaborate on voice technology applications. Additionally, there are several other universities, colleges, research institutes, and language-related NGOs. Voice technology is even recognized by the province as a crucial domain of regional interest, indicating strong political support for this master's program.

Leeuwarden is a medium-sized vibrant student city and the capital of the province of Fryslân. It is a highly cultural city, recently named the European Capital of Culture. It offers many festivals, bars, cinemas, museums, theatres, and parks. Leeuwarden, like all of Fryslân, is multilingual with many locals speaking the Frisian language alongside Dutch, English, and others. It is, then, an interesting test-bed for voice technology developments involving less-resourced languages, multilingualism, and code-switching.

Affordable housing and accommodation are easily found. The inter-city high-speed trains provide easy connection to the nearby city of Groningen and many other cities, from The Hague to Amsterdam.

Research

The MSc. Voice Technology is aligned with research activities at the faculty. Most notably this includes the “Culture, Language and Technology” research flagship. This flagship consists of an international, multidisciplinary team of Ph.D. students and postdoctoral researchers working on topics as diverse as voice-based disease recognition, multilingualism and the brain, voice synthesis for under-resourced languages, multilingual ethnography, and more.

The curriculum is aligned across four blocks.

1. The courses in this block lay the foundations for the MSc. Voice Technology with essential content needed for the rest of the program. This includes an introductory course in voice technology (in which you will develop your first application). Students also learn Python programming for voice technology and become familiar with the scientific study of speech sounds.
2. This block provides subject-specific courses on voice technology. The knowledge and skills learned in the 1st block are used together with Machine Learning within the courses Speech Recognition I and Speech Synthesis I, both of which start from the humble beginnings of voice technology, and give students an inside, hands-on view on some of the first attempts to automatically recognize and synthesize voice up to current systems.
3. Speech recognition and synthesis are taken to a higher level in Speech Recognition II and Speech Synthesis II as students learn about state-of-the-art systems and real-world challenges, augmented with guest lectures given by industrial partners. The Thesis Design course educates students on experimental protocol and research ethics and guides them through the development of a pilot study and prototype demonstrator.
4. Students combine knowledge and skills from the previous blocks and use them in their Thesis Project. Building on the pilot study and prototype developed in the last block, students write their thesis and polish their prototype to create a demonstrator suitable for their thesis defense. They also prepare to present results at the end of the year.

Courses

• Introduction to Voice Technology (5 EC)
• Programming (5 EC)
• Speech Sounds (5 EC)
• Machine Learning (5 EC)
• Speech Recognition I (5 EC)
• Speech Synthesis I (5 EC)
• Speech Recognition II (5 EC)
• Speech Synthesis II (5 EC)
• Thesis Design (5 EC)
• Thesis Project (15 EC)

Study abroad

Study abroad is unaccommodated

Job prospects

Throughout the program, we invite several guest speakers from the industry. That way students will get a good impression of the various possibilities ranging from developing speech recognizers, synthetic voices, automatic detection of pathologies from speech, the use of voice technology in robotics and/or dialogue systems, etc. The voice technology industry is a fast-growing market that provides lots of opportunities for graduates.

Job examples

• Voice Forensic Specialist
• Speech Analyst
• Research Engineer
• Entrepreneur
• Ph.D. Researcher
• Speech Scientist