As the sun sinks into the Bay Of Biscay during the last days of summer in San Sebastián, Cécile Schott is playing tour guide. We're on the route of the Camino Del Norte, where for centuries pilgrims walked Spain's northern coastline. She points an outstretched finger toward a giant statue of Christ catching the light atop a peninsula crowned by trees. Then she gestures at a pile of grey boulders half submerged along the shoreline, placed to protect multi-million Euro properties from the waves—she tells me how the seas would rise up and shatter their windows, blowing the rain and sea water indoors. Later we stop in the middle of the Kursaal Bridge, where she identifies an aquatic bird drifting slowly on the ocean breeze.

I can't help but recall the scene described in "Summer Night (Bat Song)" from Schott's superb new album, A Flame My Love, A Frequency. A bat nearly flies in her window. It's so close, she can hear its wings beating. Night is falling and her mind drifts to a house, a room, another time and place, a childhood setting. Making music as Colleen, she takes everyday details, stretches them out and places them under a magnifying glass. The mundane becomes unreal and full of potential, but it's punctured by loss and absence. Connections between nature, time and the mind are blurred, almost interchangeable.

Schott's been making music for more than 15 years, slowly building a small but loyal fanbase. Across seven albums, she's used a huge range of instruments, including zither, cello, viola da gamba, music boxes, wind chimes, melodica, clarinet, thumb piano, loop pedals and more besides. No matter the instrumentation, her minimalistic pieces are honest and intimate. But as she became more successful in the 2000s, she fell out of love with being a musician, only to return years later with an entirely new approach.

Schott taught herself to sing and developed a unique way of playing the viola da gamba, a stringed instrument dating back to the Middle Ages. Despite her penchant for digitally-edited acoustic sound, A Flame My Love, A Frequency is her first record made entirely with electronic instruments. For many, it will be an introduction to Colleen, though her back catalogue is full of riches. With each new album, she rethinks her creative process and takes risks. In the past, she scoured musical instrument museums and junk shops around Europe for instruments that might inspire a new approach. When she invited me to her studio in San Sebastián, I was struck by how she gets such deep music out of such minimal means. Over the course of an afternoon, she explained how she made one of 2017's best records using just a few pieces of equipment.

How did you come across this space?

In 2010, I knew I wanted to move to San Sebastián. Before that, I had been recording in my living room in Paris. I never had an issue with noise complaints but I felt a little constrained working there. In the last few months I was there, I wanted to record my voice for the first time. I felt so self-conscious thinking that my neighbours could hear me learning to sing. Here, no one can hear me, and it's liberating to have a space specifically for your own music.

Real estate is very expensive in San Sebastián but there's still a lot of vacant spaces in this part of town. This space was in a pretty bad state but it was cheap by San Sebastián standards. Apparently this part of town was built on top of sand dunes. I learned later that this room was used to store olives and peppers in brine. Then they'd be distributed to the local bars.

You mentioned how, a few years ago, you decided to start singing. Your music previously had been purely instrumental. How was it building up your voice?

I only sang as a teenager in the bathroom when my parents weren't home. So, yes, I didn't know how to sing at all. Like a lot of people these days, I went on the internet and found some tutorials for regulating your breathing. See this notepad in front of you? There's a list of exercises that I still use to warm up. Still, it took almost a year to really get my voice out and stop singing from the throat.

Your music took a major turn when you started singing. Does there need to be an element of risk involved for you to feel like an idea is worth pursuing?

It depends on the record. When I found the sound for Captain Of None, I was very confident that it would be a good record. For my latest album, it was more difficult because I decided to do it with only electronic instruments. I was aware of the risk of losing the essence of my sound, which was based on mixing acoustic instruments with electronic processing. I knew I had to be careful because there are a lot of people making purely electronic music. So the question was, How do I retain my specificity? How do I make sure it's as adventurous as my previous work?

You said there, "When I found the sound…" Can you describe the period of research and experimentation before you decide to make an album?

I always do lots of recording tests. I've got a notebook here that documents some of the tests for Captain Of None. There's maybe 15 different combinations of pick ups, microphones and signal chains that I went through before finding the best one. It was the same process with the new record. This stage lasts a couple of months and two, maybe three tracks will be starting to take shape. That's when the real work begins. But before then, this whole testing process has to happen.

So this is specific to engineering and instrumentation rather than songwriting?

Nowadays, I can't separate the sound from the songwriting. Sure, you can take the skeleton of a song and try it with different instruments. But for me, the way I play a specific instrument and the way I produce and record it is essential to the final character of the song. Even something simple like adding delay can cause an important shift in direction. Moog's MF-104M delay has been important for the new record in that respect. It was the same with the Moog MIDI MuRF. The way it generates filtered rhythms was particularly crucial in making the music what it was. So I'm experimenting with the sound while I'm writing in the hope that I'll find something melodically original, knowing meanwhile that the machines are influencing how the song comes together. I hope that they will help the song become what it needs to be.

This reminds me of something that struck me while watching videos of you practicing online. Some of the key moments in the new tracks are caused by simply changing a certain frequency or envelope on the MIDI MuRF. You alter the settings in such a way that it signals a new section or shade of feeling.

I know the part you're talking about. The change in frequencies causes a new note to stick out, which gives things a stronger rhythm. It came out of improvising. I would have done it for the first time and quickly written a note about it so I could recreate the effect. I have tons of notes like these. Now that the album is done, I have to rehearse for the live shows, so there's this process of going back through the notes and trying to understand what they're referring to. For instance, there'd be cues for effects settings and changes I need to make during each song. I have these enlarged blueprints of the Moog pedals that I use to write down the specific settings for every track. If the settings aren't exactly right, the song doesn't work properly.

Another key ingredient in the new record is the Critter & Guitari synths. They're quite unusual as far as modern synth designs go. What drew you to these?

King Britt contacted me a few years ago. I didn't know he was a fan so I checked out his music and watched a video of him in his studio. That's where I first saw the Moogerfoogers and I got the MF-104M after that. Then in 2015, he organised a show in Philadelphia for me. While visiting his studio, the Pocket Piano caught my eye because of its size, so he showed it to me briefly and I was immediately convinced I should get one.

Initially, I wanted something very small and light that generated simple sine waves. The idea was that I'd use the Moog pedals to cut frequencies and create rhythms and add the viola da gamba on top. You know those rhythms that you hear on old organs? That's what I was aiming for. So I tried the Pocket Piano and the MIDI MuRf, which I had tested both at King Britt's studio and with the French musician Dominique Grimaud. But I realised this rhythmic idea wasn't happening. Instead, I generated melodic ideas that had implied rhythmical elements that I really liked, which gave me the direction that I ended up pursuing.

The synths and the two pedals seem to generate different types of pulses simultaneously, which gives it this bubbly, rippling quality. It almost reminds me of American minimalism. The build up of rhythms, the harmonic structure, the way adding and subtracting frequencies determines the structure…

Well, I used to listen to a lot of Terry Riley and Steve Reich. It's definitely like water. We live by the sea, and the water is a big influence on what I do. It's all generated by the synths running through the MIDI MuRF and the MF-104M. There's nothing else. It's about the interaction between the envelope and frequencies on the MIDI MuRF in relation to the delay time and feedback on the MF-104M. There are specific combinations of settings that create this bubbling you're talking about.

What role did the Critter & Guitari synths play?

I think they really come into their own with the Moog pedals. With both the Pocket Piano and the Septavox, which I bought later, I used the two octave arpeggiator, vibrato and octave cascade modes. I consciously made sure not to reuse the same settings, and likewise with the MIDIMuRF. The setup is small but powerful if you use all the different combinations available.

Your first record, Everyone Alive Wants Answers, feels connected to your later work but it was made purely with samples. Can you talk about how it came together?

It was made on my first computer, which I bought to write my masters. I learned how to type in England in the late '90s. England was a bit more technologically advanced than France and everyone was writing essays on computers already. Back then I was like, "Where is the on button?" But I had to get my own eventually. It had the tower hard drive and the giant, deep screen.

I think I've always been… not a loner in the social sense of the term but I've always liked working on my own. I had been making my music on a four-track tape recorder and a guitar. I'd slow things down and make them play backwards. Software would eventually make this much, much easier for me. Around 2000, a friend of mine gave me a version of the Acid production software and showed me the basics. It was very easy to manipulate sound in the way I wanted.

Around that time, I started going to Paris's music libraries two or three times a week. I think I was frustrated by the fact that I came from a very small town before the internet. There wasn't a music library or a music shop. I didn't have enough money to buy records. So when I arrived in Paris in 1999, I devoured the music libraries. I discovered so much music that ended up being important to me—jazz, film music, Indonesian and African music, Baroque and contemporary composition, Autechre. It was all up for grabs.

I would take tiny snippets out of the pieces that I found particularly beautiful. They'd usually be at the beginning or end of a song. And I'd make a collage out of it. I was really addicted to the process.

Did you know which samples would work well together as you sampled them? Or did you need to begin assembling them in the computer to know what was working?

I remember that I'd come home from the library, lie down in bed and just listen. I had a pen and paper by my side for taking notes whenever something caught my ear. Then I'd sample those excerpts on the computer and group them into folders. If I remember correctly, I'd only use samples taken from a single track rather than mixing and matching audio taken from multiple different sources.

Then I'd trim and cut the samples in Sound Forge before listening to them as loops. The way I was using the samples meant they needed to work as loops rather than one-shot sounds. Next I'd place the edited samples in Acid and build things from there. I'd make melodies by re-pitching the samples. Sometimes I would let the new pitch change the length of the sample and other times I would retain the original timing.

Sometimes I have a specific idea in my head but I think music largely happens as you make it. That's the way it is.

You've said before that reading about the American avant-garde composer Lou Harrison inspired you to move away from sampling.

When I began the second album, working with samples felt really stale. It's funny how you know instinctively when you're trying to repeat a past experience but it's not working anymore. Harrison wrote a lot of music for instruments he built himself. Apparently he would go to junk yards and use parts of old cars as percussion instruments. I found it inspiring to hear about someone who is getting their sounds through upcycling. That's what I wanted to do. I wanted to be the origin of the sound source.

Is this when you started going to instrument museums?

I was kind of obsessed with instruments by this point. Anywhere I went, I was going to these museums, second hand stores and junk shops. Because I was initially sampling sounds that I couldn't make myself, the next step was figuring out what those instruments were. Where are they? What do they look like? Where do they come from? Can I get something similar for cheaper?

Was there a particular time period or culture whose instruments especially appealed to you?

I'd say from the Baroque period into the 19th century. There used to be very strange pianos that were built in weird shapes. I was also very interested in what you might call proto-electronic instruments, like the Ondes Martenot. France is quite linked to this tradition. Mechanical musical instruments are also quite intriguing.

Some of your music has a miniature quality that reminds me of Pierre Bastien.

He's such an amazing person and musician. He has an innate understanding of engines. I wish I had that ability but I think it has to wait for my next life.

Although you're using an electronic setup for your current shows, you used a lot of looping and acoustic instruments on stage in the past. I think the average person might not be aware of how complex it can be working with monitoring, microphones, quiet instruments and loops on stage. Basically it's a feedback loop waiting to happen. Could you explain why it's a problem and how you got around the issue?

The background factor here is that, these days, live shows are supposed to be, or expected to be, loud. If you're headlining your own show, it's not so much of an issue because you're the focus of attention and the audience knows what to expect. But if I'm invited to a festival, it's a different story. If the act before is playing super loud, which is the case most of the time, it's impossible to compete with. When you're using microphones and loops, if just a little feedback creeps into the loop, it quickly reinforces itself and ruins the entire song. This means you can only have so much volume on stage before you get feedback. On the other hand, audiences are used to extreme volume and lots of bass. It's quite a delicate balance.

You have to put yourself in the hands of the sound engineer. On the whole, even when sound engineers do a good job, it's something of a maligned profession, if it's musicians you're talking to. I've done around 200 shows and there's been maybe two or three times when an engineer wasn't able to work with my specific needs. They try their best but the situation is also dependent on the room and the equipment. Some gigs, you have a nice sound in ten minutes. At others, it takes over an hour and it's stressful for everyone because you know you have to keep going even if the sound can only be improved slightly. But if you're using this type of electro-acoustic setup, the reality is that you know it's not always going to sound optimal. It is what it is.

What happens if you're booked at an inappropriate time at a festival?

I don't play many big festivals but there was one where I was listed to play at the same time as The Bug. I remember hearing him sound checking and I realised, "OK this is not going to happen." I'm not going to be able to hear myself and the audience won't hear anything—you might as well cancel the show. But I managed to get my set moved at the last minute.

On the whole, I'm very lucky to have a great audience. I think maybe it's because I haven't been massively successful. I never had a hit or anything. Maybe this protects you from people encountering your music randomly. Certain acts that become famous overnight sometimes have to deal with audiences that aren't so sensitive.

How do you keep your structures flexible in your live sets when you're using loop pedals?

The thing I try to avoid is an approach that's only cumulative. It's a problem inherent in looping. I use multiple loop pedals so I can sample different parts of the signal chain. It's more complex than that in reality but it means I can erase or stop specific sections. This gives you greater control over the structure compared to if you were just recording one thing on top of another. It's easy for people working with loops to get stuck in a crescendo.

Another recurring trait in your music is how you blur the line between performance and processing. There's this uncanny in-between feeling of not knowing what's played live and what's been edited.

The ironic thing about the new album is that, even though it's my first completely electronic record, it's the most live one I've made. There's no MIDI involved. I also realised that, if I fuck up a take, it's much harder to convincingly splice together electronic sound waves compared to the acoustic ones I'm used to. It's like electronic sounds are more dense and solid. If you've cut two waveforms and you're trying to reassemble them, it's very hard to do it cleanly. So there's only a couple of edits on the entire record.

On previous albums, there was a lot of editing. But when I say editing, I'm not talking about processing with effects so much as sculpting a loop until it's perfect. It's incredible when an unexpected sound gets caught in a loop and becomes crucial to the drive of a track. A small detail is picked up by the mic, then it gets amplified by the delay and becomes something quite different.

You invariably get asked about the viola da gamba. Were you specifically looking for something that could be your signature instrument?

No. It's an expensive instrument and I didn't have money to buy one for a long time. I thought, OK, this is a really amazing instrument but it's for rich people who know how to play classical instruments. Then after I started playing the cello, I saw a second hand viola da gamba in a shop, which made me realise it might actually be attainable. In late 2005, I found a luthier who built unusual instruments with integrated pickups on the bridge and commissioned a viola da gamba of my own.

It was love at first sight. It really spoke to me. Responding so strongly to a certain sound doesn't happen very often but when you hear it, you know you have to follow. And that's what I did on Les Ondes Silencieuses. This might sound like a cliche but an album is the way it needs to be at any given point. It's the music I had to make at that time. So even though it looks like my albums are quite different, they're made spontaneously and there's no calculation.

My life was quite hectic at this point. I had returned from Japan and the culture's minimalist aesthetic really spoke to me. I came back to Paris wishing my life could be like a Japanese temple. This really influenced the spareness of the music on Les Ondes Silencieuses. I don't feel so close to this album anymore, it seems very bare bones to me now. But this is how I made it at the time, and it had to be that way.

This was the period when I had to take my break from music. Too many things were happening outside of the creation process and I felt I had lost the impetus to really do things the right way. I thought I had to stop because I had promised myself that, if I was able to do music as a living, I wouldn't make an album because I had to. I was completely incapable of doing it anyway. So I took ceramic and stone carving lessons.

I had commissioned a treble viola da gamba just before I stopped making music around 2009. I felt so guilty that I had this beautiful, expensive instrument gathering dust. When I came back to music in late 2010, I was listening to a lot of music from central Asia, which has a lot of finger-picked string instruments. I really liked the sound so I applied it to the viola da gamba. Normally it's played with a bow. Then I brought the tuning down to a lower pitch and tuned the strings so the intervals between them were the same as a guitar. I knew I was on to something. In short, there was no conscious decision to develop a gimmick.

Did your experience working with other mediums eventually feed back into your music?

Stone carving taught me that the important thing is to work. My teacher started her day at eight. She worked in a very small garden in suburban Paris. No computer, no internet, nothing. I thought, This is how you get things done. Stone carving is a very slow process. You can only produce a sculpture if you dedicate days to it. This helped reconnect me to this essential prerequisite for making music. You have to put the time in, which is hard to do when you're in the flow of touring and releasing albums.