Since my previous blog post, I have added quite a lot to my piece. Following on from the research I have done on storytelling through sound, coupled with my change in theme for in the installation, I started my next session by gathering a few more gossip articles about celebrities to sample. These were:

• Hailey Bieber discussing her sex life with her husband Justin Bieber on a podcast
• Another video discussing Kim Kardashian allegedly breaking Marilyn Monroe’s dress
• Miley Cyrus apparently keeping her new boyfriend a secret to the public
• Selena Gomez announcing a break from social media over ‘eyebrow-gate’
• A TMZ report detailing ‘eyebrow-gate’

I tried to find the most vapid, uninteresting and inconsequential pieces of celebrity news I could get my hands on to really highlight the idiotic nature of this kind of reporting. I also found a sample of a military helicopter and another of an oil rig, as I mentioned in one of my previous blogs, to put into the piece and gradually bring to a crescendo and drown out the pieces of gossip as time moves along.

Now I had the samples I needed to keep me going, I began to place them bit by bit into Logic. In contrast to how I used the samples in my first session, I cut these pieces of audio up into much more smaller pieces (only a few seconds at a time) instead of letting them run for their entirety, to give a more collage-like atmosphere. I would often chop these small pieces into even smaller pieces and pan them left and right to make the listener feel as though they are being surrounded and wrapped in the drama. Below is an example of this being used in the Hailey Bieber story, which I also added some delay to in the Emergence plugin:

I used a very similar effect for the Kim Kardashian story, but without the delay:

It sounds a bit strange isolated but in the end I think it works with the underlying soundscape and other samples that are being played at the same time. On reflection though, I’d like to find a way of getting rid of the background music in the second clip as it is a little jarring in the wider mix still.

For some of the samples I added the same 1930 vinyl emulation plug-in that I used in the last session to make them sound as though they were coming from a tinny phone speaker, such as for the Selena Gomez story:

I think at this point it’s important to state that I am not necessarily trying to attack the celebrities themselves, but more the ridiculous reporting that goes with their personal lives. This was why I felt it was important to include the TMZ video as the commentary coupled with the intense background music really does feel over the top and unnecessary given the lack of gravitas of the situation being presented.

Whilst the Selena Gomez story plays out in the piece, I have placed the sound of a military helicopter that slowly comes up in volume as it comes closer to the microphone it was recorded with:

In this clip I have also put it through the Emergence granular synthesiser as I found it helped give the sound more of a sense of movement whilst also adding more of a stereo spread to it as I panned the two different streams it was being sent to in slightly different directions, which creates a larger sound – this is desirable as it needs to start drowning out the news samples.

I also put the oil rig sample in shortly after the helicopter. Initially I was going to automate it to slowly come in as it is quite a loud, industrial sound but what I found is that I actually quite like it appearing in a sudden, violent sonic movement. I think it helps hammer in the reality that if we ignore issues such as climate change and try to drown them out, we will not make them go away but instead they will come back in a more obvious, violent form where we cannot ignore them, such as climate breakdown potentially arriving within the next century. I also sent this sample through the Emergence plugin, again to give it more of a sense of movement and space as it felt a little stale without it:

I’ve kept the Justin Bieber altercation sample that I used in the first session on a loop running through the piece so far, as it has now become so drowned out that the only things the listener will hear is the shouting and swearing. This piece is telling a story, however what I’ve learnt from my research is that is best to tell this story in an abstract way to get across a mood and a point than to necessarily tell a story in a completely A to B linear fashion. This gives me the freedom to keep samples like this on a loop to help cultivate a violent and overstimulating atmosphere.

I also added a sample of a camera shutter going off, made it repeat in quick succession and panned each repetition of the clip in a slightly different direction to give an impression of being surrounded by paparazzi and also to add to the sense of overwhelm, however I’m not too sure about this as the moment as I don’t think it quite fits in and doesn’t feel as intentional as the other sounds I have used.

The piece is around 2 and a half minutes in length and is noticeably more developed than it was at the end of the last session. There are also a few other elements of noise that I have added to the track from previous experiments I had on my laptop to add to the feeling of overwhelm and dread.

Although I am happy I have developed the story more and made the piece fit in with the themes I want to cover, it is still quite rough and would definitely benefit from some additional mixing before I continue with the composition process. I also still haven’t changed the soundscape I made with the Native Instruments plugin in the first session, and I’m not sure if I will yet. I feel as though what I have worked on today is stronger than what I worked on in the first session, but that means that the start of the piece feels weaker than the mid-point. This may be ok though, as I have to realise that viewers of the final installation will not necessarily walk in at the beginning of the piece, and may walk in during the middle of the intensity. In this case, it may be good to have quieter moments with less invasive sounds, however I still feel as though I need to make what happens at the start feel more intentional so I can convey what is important.

In order to to fully realise the ideas I want to convey in my piece, it has been necessary to do some research into the wider world of storytelling through Sound Art to understand different perspectives and perhaps even fundamental truths of working with this medium.

In one my earlier blog posts I alluded to a research paper I intended to read by Mariana J. Lopez and Sandra Pauletto from the University of York titled ‘The Sound Machine: A Study in Storytelling through Sound Design’ (https://dl.acm.org/doi/pdf/10.1145/1859799.1859808). I have since got round to reading the paper and what it told me about the idea of storytelling through sound has made me further consider the kinds of techniques will use when I continue to make my piece.

The paper essentially outlines a research project in which two variations of a previously existing story (Roald Dahl’s ‘The Sound Machine’) were made into sound works – one attempted to convey the whole story through only non-verbal sound (i.e. footsteps, breath, external landscape sounds, etc.), whilst the other was more in the form of a radio drama where speech was also allowed to help develop the plot. Each version was played to a separate group of respondents who then had to explain what they think had happened in the story and recall sounds that they had heard.

One concept I found in the paper that captured my attention was a phrase originally coined by Michel Chion, which is that of ‘Vococentrism’. Vococentrism, as Chion puts it, is the idea that “There are voices, and then everything else… in every audio mix, the presence of a human voice instantly sets up a hierarchy of perception’ – what this essentially means is that when speech is used in a radio drama, for instance, it places every other part of sound design into the background as the listener pays the most attention to what is being said as there is usually more information to be gained from it.

Without knowing the term for it at the time, I think this is why I shied away from doing a radio drama when it came to starting on my piece. I felt that I really wanted to focus on building a soundscape and that having a script that runs through the piece would detract from the world-building through sound design I am attempting to achieve. Of course, I am still using clips from news reports and interviews, but these aren’t really used to tell a story in a linear fashion in the traditional format that a radio drama might, and I am layering them on top of each other to the point where a lot of what is said is sometimes unintelligible.

At the end of the paper, the researchers found that in the version of The Sound Machine with no speech “aspects of the story… were not conveyed successfully”, which led to the group of respondents for this piece not properly understanding the plot. This was mainly because the plot had elements that would be very difficult to convey without having further context as to what is happening, and it would be very difficult for someone to tell with speech or image. Then again, the respondents who listened to the version with speech also had some difficulty working out the plot, so it might have just been a poor choice of story to use for this test.

To me, what the results of the research displayed in this paper show is that if one is trying to tell a story solely through the use of sound, it should perhaps be a bit more abstract so that the general themes are easier to follow. The sound designers who created these versions of The Sound Machine were attempting to literally reconstruct the story’s plot points just using sound, which unless it is a very simple story is always going to be a challenging task without the help of visuals. This is because they were trying to develop their own idea of producing an audio film, which is essentially supposed to replicate the experience of watching a film by hearing it instead. I feel that this would be a challenging task, as there are certain plot points that will be difficult to get across without visual clues, as the researchers found for themselves. Whilst I will be working with visual cues when it comes to making my final piece, for this prototype project I want the sound to be able to stand out on its own without necessarily having the need for visuals, so that when I begin to transform it into a spatialised piece for the final installation, the concepts behind the sound are already there.

For this reason I am going to try and say what I want to say through my piece in more abstract terms, whilst still attempting to tell a story and present a narrative. I will detail my experiments with this in my next blog entry.

Since I started working on my piece, I feel that the themes I want to address have changed somewhat and now is a good time to reflect on and address this.

In my blog post reviewing Hollywood Babylon by Kenneth Anger, I mentioned how I wanted to cover themes such as the darker side of fame and mistruths that are spread by tabloid magazines about celebrities and their personal lives. I am still interested in these themes, but I feel as though the direction my piece started taking when it actually came to making it has been slightly different and it might be time to have more of a think about what I actually want to say.

I think this is partly as a result of me feeling a little stuck when thinking of ways to sonify the themes I originally wanted to cover. This led to me just trying to start with making the piece, even if I didn’t 100% know where I was going with it. This was a good way of getting my process moving, but I feel it’s produced something that’s quite different to what I may have originally envisaged.

The layering of the samples I have gathered so far into my piece seem to create a sense of overwhelm, which when combined with the low, rumbling soundscape come together to make a dread-ish mood. I like this, even though it’s not what I originally intended. I recently read an article by climate and political activist George Monbiot in The Guardian (https://www.theguardian.com/commentisfree/2016/dec/20/celebrity-corporate-machine-fame-big-business-donald-trump-kim-kardashian) that resonated with a lot of the feelings I had about celebrity gossip culture when I was searching for samples to use in my piece. In the article, Monbiot describes how celebrity culture is “an essential component of the systems that govern our lives” and mentions interesting correlations between people that follow celebrity gossip and political engagement – he mentions an International Journey of Consumer Studies report that “reveals that people who are most interested in celebrity are the least engaged in politics, the least likely to protest and the least likely to vote”. This then paints tabloid journalism as potentially a way of pacifying swathes of the population into inaction over issues with large impact such as climate change, the wars occurring in Europe and the Middle East at the moment and the future of artificial intelligence, amongst many others. One could argue that this is verging on conspiracy theory territory, however Monbiot presents an argument that is backed up with studies that confirm his hypothesis. As a species, we are currently living in one of the most potentially dangerous and volatile periods in our history as of yet, and remaining politically informed and engaged is becoming ever more important. This article was written around the same time that Donald Trump became the U.S. President, and Monbiot explains how “his celebrity became a mask for his own chaotic, outsourced and unscrupulous business empire” and how this helped him in his campaign for the presidency. I think this is true, and certainly I believe that one of the main reasons for Trump’s rise was the platform given to him by news outlets who pushed him as a wild, zany outsider to the traditional U.S. political system. Making a piece that highlights this idea of celebrity gossip culture being a mask for covering much more sinister and important topics is a theme that I believe will lend itself to storytelling through sound, and also allows me to explore a world of political engagement through sound art that I haven’t much previously explored.

Having further thought about the development of my piece, I think I will try to incorporate more violent sounds associated with current pressing political issues such as military helicopters and oil rigs as the piece goes on, with the samples of news reports eventually being drowned out by these sounds to highlight the meaninglessness of gossip culture. I would like to question the media we consume as a population and the agendas behind it. This is something I feel I can develop from the piece I have made so far and I feel inspired to continue developing it.

When I first sat down to make a start on my piece, I actually felt quite stuck for ideas. I knew the themes I wanted to talk about, but I wasn’t quite sure where to start, or what celebrity’s story to start off with.

I decided to just get stuck in with something and see if it worked, which I find is always the best way to get the creative ball rolling if I’m facing an artistic block. I found an AI text-to-speech website – Speechify, which has quite a few examples of celebrity voices which you can make say anything you put into the text box. I decided I’d just start with a random well-known celebrity that was included on there, so I typed some gibberish in and had an AI version of Gwyneth Paltrow read it back to me.

After a while of messing around with the text to speech generator, I realised that it didn’t feel like a great jumping off point for me. Although it was entertaining to use, the more and more I thought about it I realised that I didn’t really want to write a script for this piece, which is essentially what I would be doing if I were to use this technology as the main focal point of my work. It would be similar to a radio drama, a medium I have enjoyed making in the past, but that I don’t feel works with the aesthetic of what I want to present in my final installation. I would like the soundscape to be more abstract and for people to piece together the meaning from the sonic and visual elements – having AI read out stories felt a little too on the nose for me. Of course the sound in the installation should be able to stand on its own and tell a story without the visual elements being there, however I feel like there are other ways I’d rather do this.

So, after realising I didn’t want to use Speechify just yet (although I may come back to it later in the making of the piece) I decided to again just get stuck in with something else without thinking too much about it. I had the idea of working with collage to help tell the story and convey the meaning that I want. Similar to how I want to use physical collage in the form of newspaper cuttings stuck to the wall in the visual part of my installation to fabricate fake stories about celebrities and distort truth, I decided that a sound collage could also help me create some parts of the story I want to tell, whilst still being abstract enough that audiences can take away their own meaning from the piece.

I downloaded a few celebrity gossip interviews and reports and put them into Logic to see what kind of collage I could make with them. I started by placing a report of Tom Cruises’s promotional videos for Scientology into my DAW:

I also downloaded the audio from a video from 10 years ago of Justin Bieber ‘lashing out’ at a member of the paparazzi (although in my opinion in the video it looks like Bieber is being harassed by a man many years his senior, he is only 19 years old in the video – another example of the press distorting reality):

I put the Tom Cruise interview through a granular synthesiser as an experiment; I don’t really have much experience with granular technology and have never used it in any of my pieces prior to this one so I thought I would try it out and see what results I could get. I think that conceptually granular synthesis also works with what I am trying to convey with my sound collage – it chops up and messes around with the original source material put through it to come out with something different, similar to how celebrity stories can be treated by the press. I used the free ‘Emergence’ plug-in by Daniel Gergely for this process:

I ended up going for a setting that didn’t mess around with the original audio too much, as I still want the audience to hear what is being said. It ended being a delay-type effect which I have faded in and out with automation. I then put the track through the iZotope Vinyl emulation plug-in and used the ‘1930’s’ setting to remove a lot of the low-end give it a telephone EQ sound to make it sound as if it is coming from a cheap phone speaker:

This is the finished result:

A lot of what Tom Cruise says in the interview such as ‘We are the authorities on the mind’ actually comes across as rather controlling and creepy, and I think the warped sound of the vinyl plug-in plus the delay helps accentuate that. There is also a moment later on in the clip that has a crowd cheering for Tom Cruise, but with these effects put on and placed lower down in the mix amongst other tracks it begins to sound like screaming, which is again distorting the original context of the scene that originally took place when this was recorded.

I didn’t really add anything to the Justin Bieber clip and am just letting it play out for now. It sounds violent and you can only really hear the swear words come out when people shout, everything else is quite muffled.

I also found a news report about Kim Kardashian allegedly breaking Marilyn Monroe’s ‘Happy Birthday’ dress at the Met Gala and ran it through the Emergence plug-in to make it much more abstract, however I’m not sure what to do with it yet so I’ve just put it at the end of the project and will think of a more intentional use for it later on:

I felt as though the soundscape was empty so for now I have used an instrument from Spitfire Audio to fill some of the sonic space in the background. I don’t know if this is a sound I will use in the final finished version of the piece, or whether I will develop a more complex soundscape to go under the celebrity interviews, but for now I quite like the creepy drone I have underneath the piece. I have been watching Twin Peaks: The Return recently, which David Lynch not only directed but also did the sound design for. I really enjoyed the ominous atmosphere created by the sound in this television series; it’s completely different from what was used in the original run of the show in the early 1990’s. It sets a completely different tone and atmosphere for the return and plays a very important role in The Return. The sound helps move the story along, with motifs such as the vinyl crackling sound throughout the series giving away small clues as to what themes/story arcs are being represented in scenes in which it is featured. Overall, I’ve found watching the series very inspiring for my sound work in terms of not only how it helps tell the story, but also the sombre, empty atmosphere it sets up. So, for the placeholder soundscape which I may or may not keep in the actual piece, I found a sound that was inspired by some of the sounds in The Return.

I used the ‘Glass Piano: North Star’ instrument on Spitfire Audio:

Then I ran it through the Emergence plug-in again, with different settings this time:

Then I put it through the Valhalla Supermassive Plugin to give it some space:

And then I ran it through the vinyl emulator plug-in again (this time on the 1970 setting) so that I could warp the pitch and give it a glitchy crackling sound:

And this is some of the final result:

I am very happy with the sound – it is moody and atmospheric and I’m glad that I’ve really come to terms with the granular synthesiser in making it. However, I want every sound I use in the piece to be intentional and somehow involved with moving a story/narrative along. Although I like the sound, I find it hard to justify it’s place in the piece other than ‘it just sounds good’. This is why I say it’s in the piece for now but I may make a more intentional soundscape in future sessions.

Having to come to the end of my first session with the project, I am sort of happy with it but I realise I still have a way to go. Hopefully I can start to develop more of a narrative through my piece as I get further along in the process.