Wednesday, 27 May 2020

BBC Radio Devon

Just a quick post to let you know that I will be on BBC Radio Devon's Breakfast show tomorrow (Friday 29th May) at around 7.50 to talk about the soundmap project. You can listen here: (or on sounds afterwards).

Tuesday, 26 May 2020

Covid Soundmap

Over the last weekend I have completed the covid soundmap. I added the final two recordings to the map and have made a final version of Music Machine 45. The last two recordings were both made on Thursday 21st May in Exeter. The first was made in the morning when I walked into town and features the sound of some of the building work that has restarted on the bus station area. The second recording was made around lunchtime when I walked near County Hall. You can hear the sounds of children playing at the nearby St. Leonard's school. Both of these recordings provide a contrast to the earlier ones that featured bird song and little else; now the sound is dominated by human activity as the lockdown is eased.

The soundmap can be found here:

and all the audio recordings (including the first versions of Music Machine 45) can be found here:

This project was achieved with support from Kaleider.

Tuesday, 19 May 2020

Audio recordings, locations and process

Here is a little bit more information about the recordings and the process that is gone through to achieve the final audio collage.

I have made the recordings in locations that I have visited during this lockdown period either on foot or by bike. I intend to make a total of ten of these recordings.

The locations were not been chosen in advance; I take my recorder with me and make a recording when I find a suitable spot. However, I have found that my approach to making recordings has changed over the past few sessions. Initially I was happy to record at any location on my journey; I have found that as the project had progressed I actively seek locations that have more interest. So, for example, today I have made two recordings and have decided to use both as they have interesting elements within them.

The first recording I made today was close to an electricity sub station; you can hear the hum from that throughout the clip. The second recording uses my home-made hydrophone that I placed in a small ford in Thorverton. The sound changed quite noticeably half way through this recording as the microphone moved in the current.

The longest clip so far is the one that I made on Saturday in Starcross. I have kept the full recording as there are many interesting sounds on it. There is part of a conversation that two people had on opposite sides of the road, there is the sound of a train announcer, the train itself and a pelican crossing. Because many of the recordings have been dominated by bird song this one felt quite different.

The process that happens once I have all the recordings is as follows. The code for Music Machine 45, in common with most of the machines, is written in python. It is a digital Musique Concrète maker (or should that be mixer?) I can put in as many source recordings as I wish; the machine then decides how many smaller clips it is going to make from the originals. It then generates those smaller clips and decides how long they will be and from which section of the original. Once the clips are generated they are played back at various intervals, faded in and / or out, set to sound at a particular volume and location in the stereo field. As I say, this is very similar to Musique Concrète techniques and produces a work that is multi-layered and built from existing recordings.

Here's the link to the soundmap:

You can listen to all the recordings (including the early finished versions of Music Machine 45) here:

Thursday, 14 May 2020


I was fortunate to receive some funding from the Kaleider Kfund and have been working on a project that uses audio recordings to create a new piece.

My original intention was to create a music machine piece from some audio recordings made whilst we have been in lockdown. This was partly inspired by the Silent Cities project that has asked people to record the sound of their cities during the lockdown. It's described as 'a participatory monitoring programme of an exceptional modification of urban soundscapes during Covid-19 containment'. A similar project that I have contributed to is Pete Stollery's sound map which aims to capture sonic environments which have changed as a result of governments’ actions around the world to curb the spread of the virus.

As with all projects that have a relatively vague starting point the project has changed a little since I started it. I decided that rather than use recordings from my local neighbourhood that I would take my recorder with me when I went on my daily exercise (either walk or cycle). I have now made a soundmap that shows the locations where I have made recordings and it will play those recordings when the marker is clicked. It can be found here:

or via this shortened link:

The orange markers are the sites where I have made recordings. You can hear a lot of birdsong, which is not too surprising considering some of the locations. There is very little human noise; some cars and motorbikes but no conversation. The two recordings that have interested me most are recording 3 and recording 5. Recording 3 was made in a location very close to both the airport and the A30. There is, of course, no aeroplane noise and the traffic is much quieter then usual. Recording 5 was made in very windy conditions, some of which you can hear. The recorder was placed near to a metal gate and, at around 3.38 in, you can hear the gate 'whistling'.

I have written some new software that will manipulate these recordings and then play them back. The machine will edit the full length audio clips into many shorter ones and will add a fade in or fade out to them as well. Once this processing is complete the machine will start playing back the audio files at random times and volumes to create a new soundscape. This piece will change as I add new recordings. The most up to date version can be found by clicking the blue marker (numbered 0) in the centre of Exeter.

I plan to make this project encompass a period of around ten days. It will be interesting to hear how the sound changes as the rules around lockdown are relaxed. In common with many people I have noticed how much I can hear the birds at present and how little traffic there is.

Saturday, 2 May 2020

Activity during lockdown

Having spent the past few weeks in lockdown has given me a chance to review some things and catch up on others that I have been intending to do but have previously not got round to.

It was extremely disappointing to cancel the Music Machine exhibition in March. Music Machine 4 for multiplayer was also to have been included in the Game>Play festival that was planned for the summer; that too has now been cancelled.

My current plan for Music Machine 4 is, once it is safe to do so, to make a video of eight people performing it.

As for other projects, today I have set up a live stream that is on the locusonus sound map and this is a part of the soundcamp world dawn chorus day.

I have been making some other audio recordings using my Zoom h2 and its new windshield:

These recordings are for this Silent Cities project that has asked people to record the sound of their cities during the lockdown. It's described as 'a participatory monitoring programme of an exceptional modification of urban soundscapes during Covid-19 containment'.

I have also made a new recording of Music Machine 29 which you can listen to on soundcloud or bandcamp (where you can also buy it). This was partly inspired by Robert Fripp  releasing his Music for Quiet Moments (described as
Hopefully something that will nourish us, and help us through these Uncertain Times.)

Music Machine 29 was composed for the Sonorities Festival 2014. The theme of the festival was 'Remembering and Forgetting'; Music Machine 29 works in the following way:
Music Machine 29 consists of two parts. The first part plays a C major scale (both ascending and descending) and arpeggio. The second part remembers each note that the first part plays and stores them in a list. It will then choose a note from that list to play back simultaneously with the first part. Occasionally the program will delete (forget) some notes from the list and then start remembering again. There is also the chance that notes will be sustained allowing chords to build up.

The piece is performed using a midi wind controller which is played through a Raspberry Pi and it is the Pi that does all the processing work. In the past I have used wind instrument sounds, usually flutes, but for this version I have used strings. I have also extended the piece - have a listen and see what you think!

Music Machine 29 (2020 edition) on bandcamp

Music Machine 29 (2020 edition) on soundcloud