In our ongoing AI in Your Community series, I spoke with Emilia Gómez, a researcher on music information retrieval—an interdisciplinary area dealing with music and artificial intelligence. She works at the Music Technology Group, Universitat Pompeu Fabra in Barcelona, Spain.

Tara Chklovski: What problem are you currently working on?

Emilia Gómez: My research deals with music. We try to understand how humans describe music (e.g. which instruments are playing which kinds of music, if a song is sad or happy), and we develop algorithms that can perform these descriptions in an automatic way. These algorithms are used, for instance, to search for music in a very large collection (e.g., find a song with a guitar, find a song with a fast tempo, find a piece that is similar to this other piece). This is then used to learn about music, compose music with algorithms, or recommend music to people according to the music they listen to.

Emilia Gómez

Emilia Gómez

The last project I worked on, PHENICX, used music technology to enhance the way we experience classical music concerts.

TC: What inspires you?

EG: My inspiration is to create technology that can help people interact with music and that can make all music accessible to everyone.

TC: How do you think AI will help strengthen society and communities?

EG: Artificial Intelligence can be used in many different ways, so it is important to create AI tools that can assist in different tasks, complement our skills, and increase our opportunities. It is also important to understand how artificial intelligence works, and the secrets and good practices that should be followed.

TC: What aspect of AI excites you?

EG: I enjoy working with data and discovering certain patterns that cannot be seen on a first sight. For instance, in music, we can use AI to analyze lots of data about music and discover which are the characteristic melodic patterns, for instance, of a certain composer. This helps us to get another perspective on music than what can be achieved by listening.

TC: What makes a good problem?

EG: A good problem needs to be relevant, easy to explain, and easy to formalize, which means specify the needs and divide it into small pieces or parts.

TC: What advice would you give about designing a good product?

EG: I am an academic researcher, so I am not really designing products. But as a researcher I think any product should be intuitive to use, personalized to each person, and made to work well.

TC: What do you find difficult and how do you overcome that difficulty?

EG: A main difficulty of working in an interdisciplinary environment is that you need to combine knowledge and methods from different areas, and this is complex to achieve. It also is a challenge to be a female engineer and researcher, as we are a minority, yet on the technological landscape. In this respect, I enjoy contributing to Women in Music Information Retrieval (WiMIR), a group of people dedicated to promoting the role of, and increasing opportunities for, women in the music information retrieval research field. The group meets to socialize, share information, and discuss in an informal setting, with the goal of building a community around women in our field. There is also a mentoring program where we support girls and any women interested in music and technology.

TC: What advice would you give children and families as they try to find a problem in their communities to solve using technology?

EG: Find a problem that is relevant to you, as it is important to have the motivation to work hard in a problem. In my case, I love music, so music is the driving force for me to get into technology and artificial intelligence and to apply these techniques to help people listen, learn, and interact with music.

TC: What do you think is the best way for children to learn more about AI?

EG: There are ways to learn about the basics of artificial intelligence. For one, we can start by learning about how AI is applied in the applications we use everyday in our mobile phone or computer. Then, we can also learn how AI is achieved with computer programming. I recommend this interesting blog post for kids!