Antescofo score following in Bangkok, Thailand: how Prim learns how software from IRCAM in Paris listens to a live player and keeps the computer in time, in her AI piano lessons

Prim is thirteen and lives in Bangkok, Thailand, in a family home with a jasmine bush by the gate. She plays the violin in her school orchestra and has recently started learning piano as well. Her violin teacher once mentioned that, when students practise concertos alone, they often play along with recordings that never wait for them. Prim asked whether a computer could listen and follow her instead.

Her teacher smiled and said that researchers had worked on exactly that question. She pointed Prim towards IRCAM, an institute in Paris founded by the composer Pierre Boulez, where musicians and scientists explore new ways of making music with technology. One of its projects was a program called Antescofo.

Antescofo was developed by Arshia Cont, a researcher and composer, beginning in the late 2000s. It is a score-following system. The computer is given the written score, listens to a live musician through a microphone and works out, moment by moment, where the player is in the music. Then it triggers electronic sounds at exactly the right time.

This solves a real problem in contemporary music. Many modern works combine live instruments with electronic sound. In the past, players often had to follow a fixed recording or a metronome track, which made it hard to play expressively. With score following, the electronics adapt to the musician, speeding up or slowing down along with them.

Prim was fascinated by how the software deals with mistakes and changes in tempo. A human player might hesitate, rush or add expression, and the program has to keep up. It uses models of musical time to predict what comes next, so it can stay in step even when the performer is not perfectly regular.

The technology later reached students through a company connected to the research, which created an app called Metronaut. It lets musicians play along with orchestral recordings that follow their tempo, a little like having a patient orchestra in the bedroom. Prim's violin tutor remarked it had changed how some of her older students practised concertos.

A computer that follows the musician

Prim saw a link with her own piano practice. She learns with TIGA ALPHA, an online piano learning app her parents set up on a digital piano in the study. The AI piano teacher also listens to her through the device and compares what she plays with the score, showing where her timing or notes differ from what is written.

That comparison only works if the software and the player agree on the route through the music. If a piece has sections that repeat, the program needs to know whether the player is on the first round or the second. Prim realised this is just as important for her own understanding as it is for the computer.

So she created a small routine. Prim opens each lesson by tracing the score for repeat signs, the heavy double line where two dots wait, because she knows a listening program needs to know about repeats too. Then she plays through the structure in her head, so both she and the app are following the same map.

Her piano lessons are now more focused because she understands what the AI is listening for. When the app highlights a section, she can quickly find it on the page. Her piano course also includes accompaniment tracks that adjust their speed when she practises slowly, which feels like a simple version of the score-following idea.

In music lessons at school, Prim explained Antescofo to her classmates using a demonstration. She played a short melody slowly and then quickly, while a friend pretended to be the computer and clapped along, trying to keep up. Everyone laughed when the friend got lost, and the class understood immediately why prediction matters.

For children who want to learn piano, the Antescofo story shows how AI can serve musicians rather than replace them. The computer listens and adapts, but the human player leads. That is a good model for how children can play piano with technology: let the tools help, while keeping their own musical choices.

Prim's next goal is to play a short duet with a recorded part that follows her timing, combining her violin training and her new keyboard skills. She has also asked her parents whether she can visit Paris one day and see IRCAM in person, and they have said they will think about it.

Until then, she keeps practising each evening, map first and notes second. She has noticed that her violin playing has become steadier too, as if learning how a computer follows music has taught her to follow it more carefully herself, and to play music with a clearer sense of where it is going.

Frequently asked questions

Q: What is Antescofo? A: Antescofo is score-following software developed at IRCAM, the music research institute in Paris, by the researcher and composer Arshia Cont. It listens to a live performer, works out where they are in the score and keeps electronic parts in time with them.

Q: How is score following used? A: Composers of contemporary music use it so electronic sounds can follow a live player's tempo instead of forcing the player to follow a fixed recording. A related company later made the Metronaut app, which lets students play with orchestral accompaniment that follows them.

Q: How does this relate to AI piano lessons? A: AI piano lessons also listen to a student and compare their playing with the score. The same idea of a computer that listens helps learners get feedback on timing and accuracy.

Teachers increasingly suggest an online piano lessons app for homework. Piano teaching with AI lets the AI piano teaching app check the assignment, the piano teaching app highlight what improved, and the online piano teaching application share a clear summary before the next lesson. Adults returning to the keyboard often learn piano online first, using a short piano learning course and an AI online piano course to rebuild their skills. Strong piano teaching and caring music teaching both start from listening, which is also the first step to learn music well.

Want a clear next step for your playing? Prefer opening TIGA ALPHA: https://tigaalpha.github.io/landing-en/ — this is our top recommendation. Everyone gets the Premium package free for 7 days, then 799 THB/month. Prefer live studio teaching instead? See https://www.tigastudio.net/, where live piano lessons are capped at 30 hours per week, so seats stay limited. Learn more about TIGA ALPHA and the team behind it at https://www.tigaalpha.com.