Here’s the trailer for Pessoa, the multi-dimensional, integrated audio description and captioned short film I made through an Arts Council England-funded Develop Your Creative Practice project.
My DYCP research generated so many ideas and options for AD. When it came time to create the descriptions used by Robert in the film, I was guided by choices that reflect how he, as the describer, would respond to the experimental film – which already has a voice in it – my voice!
The idea for Pessoa started with the image of someone living alone during lockdown, a writer like Fernando Pessoa with his multiple personalities. I haven’t appeared in one of my own films since 2007, but in this filmI was able reflect myself multiple times, as the auteur whose work Robert describes, and again in the outer layer of the soundtrack, describing Robert and Violet and the others. My desire/ability to CONTROL EVERYTHING reaching new heights, exasperated and exacerbated by Covid-19, 20, 21 and 22!
This film was created by me, the actors – three in their own homes and one in my studio – and two other people, Zeynep Dağlı and Çağlar Kimyoncu.
“Multiple soundtracks, that’s definitely the way to go!”
For part 2 of my Develop Your Creative Practice project, I chose to rework the conventional audio description that I had previously created for two of my non-narrative films, flow2 and flying & floating. I had screened these films with AD, and received an overall positive response from blind and visually impaired people, but I still felt that the standard AD didn’t quite do the trick. Because there’s so much to describe, especially with quick cuts one after another, it felt like auditors were getting wall-to-wall description and never given a chance to relax into the mood and texture of the films. Soooo… I wanted to see whether some of the techniques, strategies and ideas I had come up with in my research could be applied to more interesting modes of description for these films.
Working with Calum
I have been very lucky to work with Sound Designer Calum Perrin. Calum is a sound and installation artist and composer working in radio, theatre and visual art – I first met him when he completed a filmpro online commission, Texture Recipes #1-3, translating domestic environments into printmaking images and audio textures.
Calum was keen to apply his creativity to audio description and it felt like we were equally excited about the possibilities of this work. We started talking about different tools that might be appropriate to focus on for each film.
flow2 is a succession of video images recorded in my travels around the UK and abroad. For me, as with many of my non-narrative films, its creation was intuitive. When at one point Louise Fryer asked me, “what do you want to convey with this film?” I had to stop and think. It’s about travel and fluidity, the contrast of bustle and restfulness in natural and manmade environments. But it’s also about a feeling of stop and go, of giving the audience a lot of stimulation and then space to relax, feel, be.
The first strategy I wanted to explore was to emphasize the ambient sound in each clip. In the original, the music often drowns out any location sound. Also, some of the ambient sound that I originally recorded was not that effective. Calum substituted more vivid audio of rivers rushing, birds tweeting, boats passing, and I thought, “Why don’t I always do this?”
Secondly, I was really interested in altering my describer voice to vary the sound of the narration as it was appropriate to each setting. Calum played with settings and filters to create interesting effects that a) differentiate one clip from another and b) reflect the tonal quality of that environment.
In a further tactic to differentiate the clips, Calum created a pulse or percussion sound each time the clip changes. This is subtle and I think works on a subconscious level as much as a conscious one, but adds an important non-verbal indicator that emphasizes the rhythm of the film.
I pondered writing the AD in poetic form, or adding personal reminiscences about what that year of heavy travel had been like. But actually that felt like a whole other film. In the end, we came up with the idea of a succession of single words, spoken sporadically through the film, that would give a sense of the material quality of the film, the textures and movements. When I was thinking about captioning the film, I decided to add a graphic for each of these words.
We also pared down the music. You hear it once at the beginning and once at the end, and then at other points you only hear echoes or traces of it, separate stems from the original track. This allows for more of the ambient sound to come through.
As for the description, hearing it again after several years, I tweaked the language, making it more active, and shifted the emphasis in some for some of the clips.
Here are the two versions of flow2, the first with the original AD, and the second with the reworked AD. (Captions are available.) What do you think?
flying & floating
For flying & floating, I wanted to explore a more radical approach, where the form of the description better suited the collage nature of the video. Pablo Romero-Fresco had played me a clip of some AD prepared by one of his students, a sung version of the audio description for Walt Disney’s Fantasia! Inspired by this, I wrote words and melodies for each of the images in flying & floating. Some of them use a subjective voice. For example, instead of describing the burnt-out Brighton pier as an observer, I sing in the voice of the pier itself.
Without saying “split screen”, I wanted to convey the split-screen structure of this video, where two images play simultaneously, reflecting and echoing off each other. We mixed the audio so that the tracks are panned left and right. Mostly, the focus alternates left and right but at times the two sides overlap. This can a bit challenging, when your ears are being pulled in different directions at once. But hopefully we have found a satisfying balance, where this effect mirrors the visual.
Here are the two versions of flying & floating, the first with conventional AD and the second with the enhanced AD. What do you think?
Following these reworkings, I have had several conversations with AD users, playing them the original and new versions and getting feedback. I met with performer and academic Amelia Cavallo, performer/director and AD consultant Chloe Clarke, filmmaker Raina Haig and audio describer Anne Hornsby to get their reactions.
The consensus seemed to be that the new versions show improvement and give the AD listener a better and richer experience. Of course there were lots of suggestions of what could be tweaked and areas for further investigation.
Raina Haig has thought long and hard about modes of audio description, in her own film work and that of others. Raina got very excited by the single word “textures” in flow2 – it actually brought her to tears! She felt that the new soundtrack created a sense of space and dimension and helped to orientate the viewer. “It filled in that area of the film that was missing before, the patterns that went from one sequence to another, or one shot to another.”
Anne Hornsby agreed (“One word can just say so much”) but emphasized that the single words dropped in worked best where there was breathing room around them so that they didn’t clash with the rest of the AD. Anne also really liked the filtering of the AD voice, so for example, where it sounds like it’s coming over a Tannoy system, and suggested this could be used in even more of the clips.
All of my interviewees agreed that the improved ambient sound was an effective, non-verbal way to draw the audience deeper into the videos, and most would have liked even more breathing space where you would are able to relax and absorb the ambience of that spatial environment.
Amelia felt that the new version of flying & floating, while an engaging experience on its own, did not quite orient the blind and VI viewer enough to what is going on visually and would leave people wondering. They suggested that a short audio introduction would be helpful. Chloe Clarke agreed, responding that the split screen effect was not conveyed to her clearly by the new audio description. She also felt that the music pulled all of the separate images together rather than differentiating the visuals into separate clips, as they appear to sighted viewers.
This made me wonder to what extent an audio description track needs to convey the precise visual experience to the viewer, and to what extent the AD version of a film could be its own parallel experience. Once again, the solution would seem to be CHOICE. Rather than an audio intro, I would like to provide alternate AD tracks so that someone could listen to both the standard AD, giving an objective or literal description, and the enhanced approach to get a more creative, emotional take on the film.
But I’m still working out what would be the best platform for providing alternate audio tracks, other than supplying multiple versions of each film. Currently, you can’t toggle audio tracks on either Vimeo or YouTube, although there are test versions running and I suspect this feature will soon be available.
All of this work has been exciting for me, but the more I’ve explored the more I want to try. I guess the biggest points of learning have been around how much further one can go with audio, not just for audio description but in my filmmaking practice as a whole. In this part of the project, I gave myself the challenge of working within the constraints of the original edits (with one cheat!) but in future, considering the AD at the start and giving more consideration to audio throughout will affect the way I structure and edit these non-narrative films.
And options! There’s no description that is going to suit everybody, and different modes of description for the same work have the power to interact and inform each other. My next step is to research platforms that can make this more accessible. And stay tuned for my short film Pessoa, where alternate soundtracks recur as an ongoing motif!
In April 2021, I was awarded a Develop Your Creative Practice (DYCP) grant from Arts Council England (ACE) to explore the intersection of audio description (AD) and filmmaking in my artistic practice. My ambition is to create a space where I can look at innovative modes of audio description for narrative and non-narrative work, and begin to incorporate aspects of these into my own filmmaking and audio description practices.
A little background
I came to audio description organically, having worked with blind artists, describing for them in situ (workshops, rehearsals, real life), and also from audio describing my own work. It was only once I was doing this for a while that I started to audio describe work for other artists’ projects. To date, this has included independent video works, visual arts exhibitions in galleries, performance and live art, cabaret and even panto!
My own filmmaking practice includes a mix of narrative and non-narrative work. My 2016 short Awake, featured actors Alex Bulmer and Margo Cargill, who are both blind. The film was well received and won awards – the chemistry between the characters, Anna, a depressed and ill woman with a bleak sense of humour, and Doreen, an upbeat Jehovah’s Witness who won’t take no for an answer, is surprising and enjoyable. The film was seen as groundbreaking in its lack of focus on the characters’ impairments. The characters are blind but the story is about something else.
My non-narrative work comes from a place of intuition and play. Rarely do I set out with an idea of some meaning or message I want to convey. Rather, these shorts evolve through a process of discovery: looking through my video archive, choosing clips, juxtaposing them, adding some music and soundscapes and seeing what interesting connections and resonances emerge.
Interestingly, when audio describing my own work, both narrative and non-narrative, I have defaulted to quite standard and/or conventional modes. There is still something to be said for this – sometimes access can get too gimmicky and intrude on a non-sighted person’s enjoyment of and involvement in the work itself. At the very least, the goal should be to provide basic access.
With this DYCP grant, however, I wanted to examine and expand my practice. Working in consultation with writer-director-actor Alex Bulmer, my programme of work is divided into three parts: 1) in conversation with AD users, academics and describers, to look at current examples and find models of innovative practice; 2) to go back to my own previously described films and develop alternate, more creative audio descriptions and then compare them to the originals; and 3) to create a new work that combines both narrative and non-narrative elements and lends itself to a non-traditional, innovative approach that is conceptualised from the start as part of the film.
Just as I was starting the research for my project, I saw an advertisement from VocalEyes who were linking audio describers up with more experienced mentors as part of a Digital Mentor scheme, in association with the UK’s Audio Description Association (ADA). VocalEyes is one of the UK’s premiere companies rolling out audio description, advocating for widespread practice, and facilitating training for describers.
I thought this would be a great opportunity to bring an additional person on board, a sounding board with whom I could review my plans and progress. I described my goals in my application and Matthew Cock from VocalEyes linked me up with mentor Louise Fryer. I was thrilled! Louise is at the forefront of audio description research and practice. Her book, An Introduction to Audio Description – A Practical Guide, is the go-to text book for describers, a work I studied that has continually influenced my AD practice. Her recent work includes Integrated Access in Live Performance, a survey of integrated audio description (with Amelia Cavallo) in performance. Another key influence on my thinking around AD for film has been Louise’s work on the film On Blindness, which offered a standard AD track as well as an alternative, enhanced version that included additional audio material and dialogue. People in the UK can watch a short film about the process behind this here.
I had an initial Zoom conversation with Louise, in which I outlined my plans and the questions I wanted to research. I was impressed by her generosity and support as well as her flexible thinking, and Louise helped me to solidify my plans for the research and practical work ahead. Despite having written “the text book” on audio description, Louise remains curious and welcoming to new ideas and continues to explore new practices.
New Directions in Audio Description Technology
Two people Louise suggested I talk to were Ben Shirley (University of Salford) and Pablo Romero-Fresco (University of Roehampton), and she helpfully provided introductions. I followed up in Zoom sessions with each of them, after doing initial research, reading their articles in academic journals and Googling projects with which they are involved.
Researching Ben Shirley’s work and then talking with him on Zoom sent my head spinning with possibilities for creative innovations in AD. Much of his work has explored “object-based audio”, that allows audiences to choose individual levels for different audio “objects” such as dialogue, music, and background ambience, adjusting the balance of these to create one’s own optimum AD experience. In 2019, some of these techniques were incorporated into an episode of BBC’s Casualty.
Further research has explored the personalisation of AD including looking at and experimenting with factors such as: the speed and style of delivery or AD voice; the location of the AD voice relative to the action; using alternate devices (more about this below); and information richness (how much detail to include).
As I discussed with Ben, I find this area of research extremely exciting and provocative. However, that’s tempered by the relative inaccessibility of cutting edge and experimental technology for an independent filmmaker. I’m looking for strategies that I can employ in my own work, and that can be rolled out to viewers without expensive equipment or super complicated processes. Many of these innovations are still in development stages. Thinking creatively, how can they be adapted or implemented in DIY format? Immediately!
The S3A project team (researchers from the Universities of Surrey, Salford, Southampton, and BBC R&D) developed a way of “using speakers in devices that people already have at home—mobile phones, tablets, and laptops—in order to unlock immersive audio experiences for many more listeners. It’s like a surround sound system, but without the hassle of cables and expensive loudspeakers.”
By attaching your laptop, mobile, tablet etc. via bluetooth, you can get different audio tracks from different devices, and place them wherever you want within a space. I thought that this could be a great application for presenting various aspects of the audio track for blind and partially-sighted viewers; so for example, the dialogue could come from your laptop, the AD narration from your mobile, and the background or ambient audio could come from your tablet. In this way, listeners can personalise and set levels on whatever combination of tracks works for them. Although this stage of the Vostok project is finished, the software is available for people to experiment with, and I’m sure we will be seeing it incorporated into further developments in artists’ practice.
I had an equally enlightening chat with Pablo Romero-Fresco over Zoom. Pablo and Louise have worked together on Media Accessibility research and I found his papers highly stimulating and relevant to my investigations. For example, the groundbreaking Accessible Filmmaking Guide, from 2018.
Like Ben Shirley, Pablo’s research has been about offering choice to AD and/or captions users – as opposed to a “one size fits all” strategy. Pablo, like Louise, is keenly interested in alternative, innovative and creative approaches, and advocates for audio description that is conceptualised at the beginning, and integral to the production of the work, rather than tacked on at the end. Like Louise, he is excited by new directions AD can take and interested in my explorations.
An ongoing question facing people who are looking at new modes of audio description is whether the AD experience should emulate (as much as possible) the experience that sighted viewers get, or whether the access strategies (AD, captions, etc.) can provide a parallel but different, even enhanced aesthetic experience. (e.g. Notes on Blindness, above).
One artist that Pablo refers to in his research really turned me on with her use of creative captions that take their own turn, creating a witty parallel narrative to the video. This is Deaf, American artist Christine Sun Kim with her video Closer Captions (with link to video description and transcript in the biog, or here).
What would a similar approach to audio description sound like? Might a playful approach engage viewers in ways that standard AD cannot? BUT can you get too creative with the access elements to the point where they are distracting? What about that viewer who simply wants access to the work, rather than a playful commentary on it?
Enhancing Audio Description
Another project that I encountered in my research was Enhancing Audio Description (EAD), headed by Mariana Lopez at the University of York, along with Gavin Kearney.
The EAD project explored ways to make audio description more engaging and effective, incorporating creative possibilities of sound design. The team surveyed blind and partially sighted AD users before, during and after their research, to stimulate active and practical methodology. Their work culminated in reworking the audio description for a short film called Pearl (dir. Hannah Palumbo) that the team used as a case study. Rather than simply trying to provide more detailed description, the researchers implemented three very powerful techniques:
enhanced soundtrack for the film itself (additional audio cues and effects)
binaural audio (to create spatial awareness of where the sound/voice is coming from, as opposed to traditional technique which places all dialogue at the centre of the screen), and
a first person, subjective approach to the description (the I-voice).
You can read more about this project here. And you can watch Pearl, with enhanced AD including binaural sound and 1st person AD here (watch with headphones!)
I had a great Zoom with Mariana Lopez about her ongoing work, and she described the next stage of the project, Enhanced Audio Description part 2, which has just begun and which will explore these strategies further. Our discussion provided amazing food for thought as I consider what strategies to inform my own practice.
Finally, my work on this project has coincided with a series of workshops offered by VocalEyes on a series of topics for describers, a Covid-friendly CPD (continuing professional development) series. I participated in several, including one called Describing for Film that featured a presentation by Timna Fibert, an audio describer who discussed her recent project describing non-narrative art films for the Alchemy Film Festival. I found Timna’s description for these abstract and subjective films to be astute and skillful. In a subsequent Zoom conversation, I was impressed with Timna’s enthusiasm around the challenges of describing non-narrative films. Her expert description came from working directly with the filmmakers to understand their intentions and the meaning of visuals in the film. It was interesting to hear her say that, on reflection, she felt like she could have gone further. For example, she said, she could have used more poetic language, phrases rather than full sentences. It reinforced for me the idea that work in this area is an evolving and open-ended discourse.
Where Is All This Going?
So, informed by all this, I am inspired to start looking at my own work and coming up with alternative approaches to try out. I definitely feel like choice is key, that ideally you would be able to offer AD users a choice of standard approach, enhanced audio, different mixes of the various elements, etc. But how you can actually deliver alternate choices is still problematic. More complicated approaches and multiple options can be a technological barrier for users. It’s still early days for some of the approaches being developed in research environments to be rolled out to mainstream viewers. However, I do have a few ideas!
Through the past months, of course, work life has not stopped for me, and while I was engaging in this research I have also been creating audio description for some fantastic projects that have provided unexpected challenges and insights. These have included Cathy Mager’s film Sign Night, a dance film by South African artist Siphenathi Mayekiso called Echoes of Identity, and Bobby Baker’s performance Diary of A (Grand) Mother’s Experience. There seems to have been synchronicity in being invited to work on such a variety of projects that each pose their own questions and inspire me to re-examine my assumptions and methods around AD practice.
Let me know if you have any comments, insights or thoughts about all of this, and watch this space for further developments! I am currently planning and starting work on parts 2 and 3 of the DYCP project, and I plan to hold a webinar for invited audience at the end to share my research and show examples of newly created and revised works.