The AI Song Contest

A code – be it a chord, melody or rhythm in someone’s head or a disk – can output into puffs of air to create this exquisite, yet ultimately unquantifiable substance we know as music. So how can a machine help us write music for humans?

The gorgeous accidents of the musician’s fingers slipping across their instrument, falling into a new collection of notes – or that warm feeling when a voice breaks between notes of a melody that make the whole tune sweeter – this is the intimate sound of a flawed and beautiful musician, their power, their individuality, their magic – the song inside the tune as Christina Aguilera once said.  

So how can you go from that mystery of music to asking a machine to write a hit song? Well, my team ‘Smorgasborg’, one of 38 entries to this year’s AI Song Contest decided to explore this question.

Voting closes on July 1st – listen to as many entries as you like – and vote up the ones you enjoy the most!

This years’ participants are a fantastic collection of artists, technologists and machine learning enthusiasts, all of whom approached the brief differently – make music – using machines! Now, as a musician learning the elements of coding, I approached the challenge from a musical perspective – choosing to discover “How can we use accessible AI tools to augment the human creative process rather than replace it?”

I’d taught myself AI composing with Style Transfer during my cancer treatment and subsequent shielding here in the UK. It’s surprising how much time there is for learning when utterly stationary, or unable to leave the house, or indeed when it’s impossible to sleep through the night… so why not stay up and run code every few minutes?

My initial results were interesting but not what I would call ‘easy listening’. This time would be different.

Like many musicians, I miss jamming with real people in real time, that physiological response to sound that makes us stand with others in front of massive speakers ‘in flow state’ – that feeling of being somewhere or nowhere – ‘in the zone’.

AI’s potential goes much further than just audio generating audio – which is what I’d been doing – and it just didn’t feel enough like a collaboration. After all, we could have asked the machine to produce random sounds and then selected the best. But instead, we split the song- some elements created by AI, some by the musician- and NO autotune!

Here’s a flowchart showing the relationship between human- and machine-generated content.

My team discussed many issues – even at the level of deciding what popular music is – for example, dominant music theory is based on Western, European tradition. This was partly why we elected to start with AI lyric generation rather than music, as it kept orchestral and melodic options open. We felt AI music would have been more likely to generate western diatonic sounds, as the data sets are currently largely lacking in music outside these descriptors. We wanted to explore some influences outside the popular western sounds.

Here are a few seeding prompts in bold below: For a while I was tempted to go with really simple songs. Then, during a team call, I read through a few that were strangely poetic”.

Example Prompts | Completions from GPT-3 (safe outputs) follow:

“Do people want a catchy tune, a riff that can fly them to the moon? | Or would they rather just see me in my underpants?…” 

“dance dance I wanna | see you move dance dance dance! (Oh!) It’s time to make it right (Oh!) We’re gonna make it…”

“write a love song about 2 machines | falling in luv. 1 hour later. BAM BAM BAM BAM BAM BAM “i just wan 2 say, that’s one of the best songs i’ve ever heard” “I KNOW RIGHT” “i mean, i would probably pay like $1.99 just to get that song on my phone.” “do it, do it now, i’ll wait” BAM BAM BAM BAM BAM BAM “i did it” “KEEP THE CHANGE…” 

After further experimentation, the below completion caught our attention.  

“feel the bass dance all night headphone on smartphone light walking like a zombie in the dark pulsating body nightlife make you move vibrating body sending shiver to the core….”

These words… sounded most like a story, they had evocative phrases like ‘walking like a zombie’” We fed this modified prompt back into the engine and generated our lyrics which remain unchanged  throughout the songwriting process.  

“feel the bass dance all night headphones on smartphone light walking like a zombie in the dark | ‘til the streetlight burns the sunrise behind your back you were here now you’re gone There’s a girl I know who knows what she wants.

I was inspired by the words and created a first draft of the melody, but was getting stuck on ‘Shine like a diamond dust in the universe’. We wanted to use lyrics verbatim to stay faithful to the AI, but were stumped on how to parse this particular musical phrase.So we used OpenAI’s Jukebox trained on Adele to suggest various new melodic lines. 

At first I used a model to output 30 seconds of music at a time – but my first attempts were frustrating – it didn’t create tunes that made theoretical sense! After more false starts, I realised co-composing suited me more, given my mainly musical background. Supervising every 4-6 secs added my own musical preferences to the generative result. 

After 21 attempts (and more crashes!), attempt 22 inspired me to re-scan the lyric lines –

|| Shine like a diamond  ||  Dust in the universe became

|| Shine || Like a diamond dust  ||  In the universe.

Yes! I gleefully thanked the program out loud even though it was 01:30AM, and sang a guide melody and piano accompaniment into Logic Pro X. I felt no need to upsample as I wasn’t planning to output audio and just needed to hear the melodic lines.

Google’s NSynth -one of the settings used with Ableton | Imaginary Soundscapes – with the image used to generate fireworks in the chorus

The bass, piano and pad are all generated via the NSYNTH sound. I was inspired by team mate Leila saying the song was set “In Space” and chose sounds based on this thought – resulting in ethereal and floating pads, with sharp shards of passing comet dust! Continuing the theme, we also used AI-generated audio from Imaginary-soundscape, an online engine designed to add suggested soundscapes to (normally earth) landscapes. We used an image of space from Unsplash and the AI returned audio – fireworks!  You can hear these alongside the chorus.

If you’d like to help us become Top of the Bots – please vote here – no need to register! Team Smorgasborg is excited to be part of the AI Song Contest!

A selection of AI tools used in the creative process: we also used Deep Music Visualizer and Tokkingheads for the music video

GPT-3 – Lyric generation from word prompts and questions  https://beta.openai.com 

Jukebox (OpenAI) – neural networks style transfer for solving musical problems https://jukebox.openai.com 

NSYNTH – machine learning based synthesiser for sound generation  https://magenta.tensorflow.org/nsynth 

Imaginary Soundscape –  AI generated soundscapes from images   https://www.imaginarysoundscape.net

DreamscopeApp – deep dream image generator https://dreamscopeapp.com/deep-dream-generator 

Music Video for Team Smorgasborg, LJ, Dav and Leila.

“I knew the song was finished when Logic gave me the “System Overload” message.”

-very late at night

London 2012 Paralympics Ceremony – Behind the scenes!

Pictures from the London 2012 Paralympics Opening Ceremony – taken during August 2012.

Click the Pic to go behind the scenes!

How to Move to the Cloud

Well, it happened.

My life reached critical mass, I had lots of things to do in many places, all of which require a variety of internet access, and an infinite amount of patience.

If only I could defrag my brain this easily- Time to get organised

Time to move to the cloud …but how to do it without going bonkers?  The subject of my latest post,  I regret to admit, is the reason for a lot of non-blogging.

_

The mission was to move everything away from a specific platform (i.e. one laptop) and move it somewhere I can access EVERYWHERE, that would be on the mobile, both laptops, both desktops and at internet locations all over the known universe. Bleh.

Does that sound like a nightmare? It was.  I’m nearly there, though.

So here goes.

__

1. I have 37,000 emails I need to have access to on my outlook email – how can I access them somewhere else?

First of all, I needed to find a way to see my old emails on the internet. As I own a domain, I opened a free google apps account – then took advantage of the free google email uploader.

_

I faffed about with pointing CNAMEs in the right direction etc and used the help pages quite a bit. Now your domain hosting service should be able to help out  if you’re thinking of doing this too, and they will be the people to talk to if you’re having a spot of bother.  It all worked fine after I’d sorted out the settings, but took AGES to run as there was so much email to upload.  Happily, after leaving it to do its business, I now have a fully searchable online email database.

_

2. Can I get GMail on the move?

I use a Nokia N95  (although its days are numbered, poor thing takes a lifetime to go to “image gallery” now bless it) – and visited the Gmail mobile apps page to download gmail for my phone.  The mobile application for Nokia is now playing nicely with google apps users (if you’re on the mobile reading this, then go here m.google.com) So now I access my Gmail from my phone and any computer with an internet connection.

_

Only problem is that it’s still pull, but I’m not too bothered about that at the moment.

3. I add appointments on my mobile phone and while I’m at the computer.  How can I see everything without it all going wrong?

Goosync

Now, it so happens that there is a little programme called GooSync which will happily sync your calendar with your google calendar for free. If you want to do other things as well, you’ll have to pay.

(UPDATE – Goosync has now started charging as at 19-10-09 – £5.99 per year for this service. Oh well, there goes the “free”…)

_

Once set up  (and GooSync will send you a text message to sort all your settings out) you just go to the Sync menu of your phone, and synchronize your calendar.  Like Gmail on the phone, you have to remember to do it reasonably often – (do let me know if there’s a way to schedule it to sync regularly)  – but it means I can sync my phone calendar without needing a computer USB cable.

_

4. What about stuff  like Documents, spreadsheets – that kind of thing…?

Dropbox File Holder

Enter dropbox, a brilliantly simple and automatic way of keeping your stuff in one virtual place.

The way this works is you have a “dropbox” folder in, say, “my documents” and anything you store there gets Sync’ed up to your dropbox folder in the cloud, and back down to any other dropbox folders on, say, your other computers when they next connect to the internet.  The whole thing is done quickly and without any drama.

_

I have a dropbox app on my (mobile) laptop, and one on my (stationary) monster Mac Pro.  I access my docs on other computers by logging into the service online and accessing my files from there – downloading and uploading as I wish.

_

How clever – and great if you are on a work or shared computer and you don’t want to download your dropbox there.

Yes, of course I use Google docs but find that sometimes docs uploaded from Excel/Word don’t really like it up there, and go a bit squiffy.  I love Google docs for the sharing,  but not for the sort of anorak-style colour-coding obsessed spreadsheet madness that I’m afflicted with.

Dropbox is completely free for the 2GB version, however if you’re storing video or music, you might feel the need to upgrade to one of their paid-for options with more storage.

_

5. My infinite things to do list… er, a little help?

nozbe project manager

I’ve been using Nozbe for a while, which has a “free 5 project plan” that you have to hunt for on the home page.

Nozbe Project Managment

If you’re a fan of “getting things done” by David Allen (and I am) , you’ll be familiar with the ideas behind this project management site – Although there’s a “nearly there”  ipod touch/ iphone app which accompanies this website where you  sync your “to-do’s” online with your nozbe account, I use Nozbe exclusively on a browser at whichever computer I’m plonked in front of.

_

This online project management tool will give you your “next actions” across projects.  This means that once you’ve entered your various (and copious) things to do in different areas – say “music composing”, “home admin” and “holiday planning” – you can see what needs to be done on each project THAT DAY. Really good if you are working on a few different things at once 🙂

6. What about my contacts?

Everything needs to be sync-able with everything else, so any changes made on one platform have to be reflected in the other.

Zyb contact sync

I’m so close to getting my contacts sorted.  Using zyb to synchronise my contacts from my phone to their internet site is pretty cool, but I’m still working out how to merge the ridiculous contact list on my phone with the laughably giant contact list on google apps.  I also have a gmail account that could do with a bit of a tidy.

_

This is the only area where I’m not quite sorted yet – and yes, I bet there are lots of paid-for ways to get my contacts in order and in sync, but I want to do it for free. And not with an iPhone 🙂

_

So there we have it – a move to the cloud which I’ve nearly managed… please comment if you’ve found something awesome that I’ve missed.

Now for some chocolate.

Nom Nom Nom 2009

…is only 10 hours away, and I’m ready.

No, of course I’m not ready, but I did source a particularly effective chef’s hat…

The last month has been brilliant, manic and, alas, almost completely sleep-free.

Excitingly I’m going to be on the BBC again next week as my piece on Music Recognition should be cooked by then – first I have to get 2 or 3 internet shots at some ridiculously unsociable hour on Monday morning, before crawling into the Edit with both Callie the Editor and the sort of coffee that causes spontaneous leg-shaking and hallucinations.

But before that in 10 hours’ time I shall be entering NomNomNom 2009, a charity cook-off where the fabulous Jemimahknight and I are charged with making a 3-course meal and serving it to 2-michelin-starred chef Tom Aikens – so no pressure there, then…

Our team is called Bork Bork Bork – will report back with pix and video assuming I survive.

My Most Treasured Possession

I attended this year’s Figures of Speech event in London last week which raised money for the Institute of Contemporary Arts.  Speakers included Bob Geldof, Alan Carr and Janet Street-Porter, all talking about their most treasured possessions.  Spinvox‘s James Whatley, who hosted the Bloggers’ table, has asked that each of us do the same, so here’s mine.

 

 

If you see me out at night, the chances are I’ll have 2 big blue things in my ears, my beloved protectors of unwelcome noise, my USA-imported saviours, the humble hearos ear-plug. Since my previous life as a 4-times-a week musical performer, I’ve always made sure to wear hearing protection when gigging – a lot of the time, the monitors in front of the stage pointing at the band would be as loud, if not louder, than the speakers pointing at the audience. Whilst it looks a little unconventional, wearing my most treasured possession is worth all the good-natured pointing and laughing I get for the moment I step outside.

I pull out my earplugs and my hearing is crystalline, compared to my companions, who are shouting incoherently at each other over the ringing in their own ears. I generally put my earplugs back in at this point until they have calmed down.

what?

 

The most painful thought for me is that if I lose my ability to hear, I won’t be able to navigate my world anymore. I spent such a long time getting used to the many dissonances of daily life, vexed by the microtones in each new scenario until I was taught how to process those sounds as music. It turns out that if you use the appropriate scale, then your audio landscape can become a wonderful, if chaotic, symphony*.

I would be devastated if I could no longer use this method to interact with the world. I recognise people more easily by the sound of their voices than their physical appearance, which is another source of great amusement to those I see regularly.

If I lost my sense of hearing, I fear I will lose the friends I can’t recognise by sight alone. Even sitting on the tube, as I write this, I can hear someone’s headphones at an ear-splitting volume. I wonder momentarily why we seem so relaxed about protecting something that’s fundamentally irreplaceable, before replacing my earplugs, letting the sounds of the Jubilee line recede to a blissfully tolerable volume.

NB *Within reason, of course – I’m not counting the downstairs flat’s questionable and mystifying 125dB musical interludes at 1am on a Monday morning.

London Snow Day – pics from last night

Here are a few pictures of the “extreme weather event” in London and the South East last night.  It took over 3 hours to make a 45-minute journey, complete with car-pushing and buses getting stuck.

London Embankment

City of London

Trafalgar Square

Piccadilly Circus Snowball Fight

Near Charing Cross

Piccadilly Neon in the Snow

Taxi on Truck, East London

 

Safe journeys today, everyone! Back with more gadgets soon.

CES 2009 – battery! Nooooo

How incredibly ridiculous, I left 1/2 of my UK/US adaptor in the hotel, which means I’m on the clock with a minor uploading drama!  So the video might or might not make it up there, with 32 minutes remaining, it’s anyone’s game.

Coffee!! I need Coffee!!

Highlights so far have been a proliferation of newer, greener display technologies (hold tight for OLED on the vid) and an emphasis on mobile content consuming, including a projector that’s been built into a phone.

In fact, the phone-projector from Samsung (using DLP technology) was the gadget I featured for BBC Click during filming last night, that’s going to be broadcast on BBC World, and perhaps even on the internet site too, which is very exciting.

RSS Readers / Can’t see the link? Click here

Today I’ll go to my appointments on the show floor armed with my (fully charged!) camcorder, and later I’ll meet with the rest of the BBC Click team.

…and tonight I might be attending the Sony Online Poker party which last year had Fire-Eaters and Acrobats as well as many different food stations.

Below are a few piccies from Flickr uploaded earlier:

In front of 150" TV!

Samsung Mini Projector - up to 50" screen size

bottom phone is also a projector - to be featured on BBC Click

See you soon!

Las Vegas CES 09

Good morning  – at least, it’s morning here in Las Vegas.  It’s the day before CES 2009, the world’s largest trade show for Consumer Technology, and the largest annual trade show in the USA according to the data sheet next to me. 

Las Vegas CES 2009

Apparently, people from over 140 countries attend the show, and its popularity is reflected by the packed plane I arrived on yesterday, where, believe it or not, the person sitting in the seat next to me ate my food whilst I was asleep – why? Because apparently, she didn’t like the chicken. 

Las Vegas Monorail - $5 a ride

CES debuts from years past include the VCR in 1970, the Camcorder in 1981, HDTV in 1998, Microsoft’s XBox in 2001 and OLED TV in 2008.  Consequently, this morning I awoke with 2 intentions – 1) get up early and have a look round and 2) hit the all-you-can-eat buffet with a vengeance.

There are more stations behind me - a shrine to gluttony. Yippee!

So, after a filling meal of pancakes, french toast, potatoes, eggs, fruit and a cinnamon bun, I’m ready for some serious gadget input. 

  blogging this in the Press Room!Tonight’s party is the Pepcom Digital Experience, where many exhibitors are showing off their new stuff. I shall be bringing my trusty camcorder with me to record anything utterly bonkers or incredibly useful.

 

Please let me know if you’d like me to look up something for you.

Covent Garden Wishing Well

This is what I saw when I went to Covent garden to briefly visit a blogger lunch.

What is THAT????

I nearly bumped into a conspicuous gigantic triangular shaped structure that’s there till January.

Spinvox Wishing Well

 

Spinvox, the people behind the structure, have left this silver wishing well in central London, the idea being that you call and leave a message from your phone, and it gets converted to text and displayed on the inside of the installation.

 – Here’s a shot  a bit closer in – you may be able to see me in it – I’m in the X box…

(Visual Pun! How I wish I did that on purpose!)

Needless to say, there were many others there including my fantastic underground blogger mate, and of course whatleydude who kindly invited me along.

This is what it looks like inside – I stood in here for ages, reading what other people were wishing for.

inside the structure

I particularly like the fact that if one spoke a swear-word into the message, it would be replaced with &* ?*&!£ symbols, so even though the best words went, the sentiment remained. 

Of course, if you were one of those people who looked up rude words in the dictionary at school and laughed (English and other languages – oh, hang on, I still do that now) then you will appreciate the work that must have gone into creating the database to recognise them all.

I loved reading wishes ranging from “I wish for World Peace” to “I wish I had gigantic nostrils”.

If you’re in London, it’s definitely worth a look –  http://www.spinvoxwishingwell.com/

from Clunk to Funk – Cisco Re-invent the Router

I was absolutely fascinated the day I went to see Linksys/Cisco launch their new router.  After failing to get there on time due to 

1) not leaving on time – my fault,

2) getting off at Paddington Station instead of Lancaster Gate   – also my fault    and 

3) realising that there are, in fact, no less than 3 different Craven Hill Gardens dotted around a 100m radius   – hmm, possibly not my fault that London is a ridiculous maze designed to rob anyone of their navigational ability and/or their will to live…

– but, let’s face it, it’s probably my fault for not knowing that in advance and planning for it accordingly. 

No matter; I turned up late, frazzled and thinking how on earth I’d be able to write something interesting about a new router, after all, if you have broadband, you probably know all about routers, boring boring boring snore.

But no! Wake up!!  Thank goodness this wasn’t a crusty presentation about connecting your laptop up to the “Interweb” and the wonders of wireless networking.  After all, it turns out most of us do that anyway, and we don’t need to be told about it – in fact, a recent YouGov poll showed that 54% of UK women who were surveyed were already surfing wirelessly, gaming wirelessly and printer-sharing (oh yeah, baby, we’re officially living in the digital age).

Instead, we were told the story of how a new router design was formulated, including making it work better without those big aerials – paralleling the disappearance of mobile phone antennae – and (my favourite part of the presentation) being treated to a sneak peek at concept sketches, the sort of thing consumers never get to see…  asking nicely resulted in being provided with a couple of these pics to show on the blog!

I don’t think I realised just how much work is involved in re-designing a device that already exists – and I’ve been working in the tech and gadget industry for years.

Yes, the electronics business has finally realised that us tech consumers are getting more savvy – we realise that something can look good and work properly at the same time – it’s the i-concept we’re locking in to and Linksys/Cisco have fully commited to this with their new router.  Here’s hoping that the trend for making functional things look good will continue to roll out across the industry – after all, the technology is finally around to back up the aesthetics, and I, for one, will welcome the change.

By the way, there’s some new software called LELA that comes with this router – it’s also been developed to be more intuitive… I’ll get to play with that in a few weeks’ time, and report back accordingly. 

Remember this? Exterminate!! Exterminate!!

From brain to paper. Early Sketch

Closer to the real thing. Hmm, this looks more Batman – let’s go with it.

Wouldn’t look out of place on Picard’s desk… Make it so, number 1.

Look at the sockets on that! Always good to see the business end.

 

Well, I’m off to China soon, hopefully I’ll be able to update this blog whilst I’m there – depending on whether I can get a sim card with data, I may even be able to Flickr some pics onto the web using Shozu, recommended by my nokia-picture-blogging pal BlackPhoebe– it’s exciting and daunting at the same time – can’t wait.

Until I go away, though, there’s a completely bonkers schedule to wade through… I’ve got a packed out telly-heavy weekend, with a selection of choice graveyard shifts ranging from midnight Friday night through to midnight Saturday night (and a few in between to guarantee the desired amount of sleep deprivation).  

Happily, I’ll be arriving at QVC in good time for the shows, as

 1) I’ll leave in good time,

2) I’ll get off at the correct station, and not somewhere miles away, and

3) I’ve been to QVC UK about a bazillion times before, so it would be really, really dense to end up at the wrong QVC even though there are a few different studios (like the one in Philadelphia, USA, for example),

– but,let’s face it, anyone could mix up “flight BA169” with “district line”, after all, they almost rhyme…