Introducing Visualising Voice

At the beginning of 2017, I started a new project: Visualising Voice. It works alongside my research on The Baudelaire Song Project, but also as a standalone research project. And I’m very excited about it.

It all came about when I applied for a Europeana Research Grant, back in September, when I’d only just started at The University of Birmingham. The call for applications was looking for researchers working in / with Digital Humanities tools, who could make use of resources held in the Europeana Collections, a digital platform for cultural heritage, funded by the European Commission.

I have to admit that, when I applied, I hadn’t really got to grips with Europeana Collections, though I’d used a number of the resources which are held under their umbrella. Mainly, I was familiar with Gallica – digitized resources from the Bibliothèque national de la France (BnF), and had, on occasion made use of online resources from other national libraries – notably, of course, the British Library (BL) in London, but using Europeana opened up a wealth of new resources to me.

So, I was delighted – and a bit daunted, I’ll confess – to find out, just before Christmas, that I had been awarded one of three Europeana Research Awards.

I hadn’t expected to gain the funding. When I applied, I wondered whether my project was too derivative, in that it closely follows methodologies developed within the Baudelaire Song Project Team. I also wondered whether it was too ambitious, as I was going to be working with Software Developer, Tom Cowley, to create a new interface, allowing users to get involved in Digital Humanities. However, Europeana seemed to like both of these aspects of my project – it’s collaborative, digitally ambitious, but also rooted in tried and tested methodologies which fit well when applied to materials in the Europeana Collections.

I realise that I haven’t actually explained what the project aims to do. So, I’ll backtrack a little bit. Visualising Voice aims to look at what happens when we perform poetry aloud. The project works on the premise that poetry is an inherently oral form of writing – it is meant to be heard – or, at least, some of the different features of poetry, such as rhythm, rhyme are best understood aurally, rather than by looking at a piece of paper. The Europeana Collections hold quite a lot of spoken word recordings of poems by Baudelaire. Working on the Baudelaire Song Project, I was already comfortable enough working with Baudelaire to know that this was something I could happily take on.

I could never get bored of Baudelaire, nor of Mallarmé, whose work formed the basis of my thesis, though I am wary of becoming a “one-trick pony”, and I was definitely ready to branch out and expand the scope of my research. Visualising Voice gave me the perfect opportunity to continue my research, whilst looking at new texts and broadening the extent of my research output. Working with the website and development also gave me a chance of develop my tech skills, most notably working with html and javascript. While the development of the song analysis interface is very much down to Tom, and his developer associates, I have been hands-on in building the website and it’s really opened my eyes to using the web to share my research.

We know why Baudelaire had to be in the list of poets to study, but why Verlaine and Rimbaud? Verlaine featured in the corpus for a number of reasons. First and foremost, he is very frequently set to music (though arguably not as frequently as Baudelaire). The large number of song settings of Verlaine’s poetry hint at an inherent orality, at something in his work which invites performance. Secondly, recordings of Verlaine’s poetry feature heavily in the Europeana Collections, supporting my view that his work has an important performative element, and making it an easily-accessible focus for the study. Alongside Verlaine, I chose Rimbaud. Recordings of Rimbaud’s poetry often feature alongside those of Verlaine and I was surprised to find that recordings of his work featured almost as much in the Europeana Collections as Verlaine’s. Historically, the two poets have been seen as a double-act, on account of their close but often turbulent personal relationship, so it made sense, in my study, to consider both poets. Baudelaire, Verlaine and Rimbaud together are bastions of nineteenth-century French poetry, with unique and pioneering elements to their poetic output – this made them ideal candidates for study in Visualising Voice.

Looking at the work of three poets together offered an opportunity to test the methodologies developed within the Baudelaire Song Project on the work of other poets, as well as exploring how we examine and talk about features of speech, as opposed to song. I think that analysing speech brings new challenges. While setting words to music adds an additional dimension to analysis, in some ways I this is easier to analyse, as there are more layers and more obvious variations in performance. In the case of speech, the differences between performances of poetry are more subtle and, as a result, bring a different angle to our research.

So far, I have firmed up the corpus for the project, which involves looking at three different performances of the same poem. There are three poems each by Baudelaire, Verlaine and Rimbaud, so initially, we’ll be looking at nine poems, though we hope to expand this to eighteen over time. Because of the nature of the resources in the Europeana Collections, most of the recordings used are from the 1950s or early 1960s. This limitation brings both advantages and challenges. In the first instance, we’ll need to assess how much variation there is between the different performances and what the nature of those variations are. If there is little variation, then we might see this as a generational style of performance. The real challenge is that, within the Europeana Collections, at least, we don’t have access to any more modern recordings: this is why we decided to add a crowdsourcing element into the project. In the latter phases of the project, we’ll be inviting users to record and analyse their own performances – either snippets or whole poems, as they wish, in French or in English. These can either be deleted or, if users are interested in contributing further to the research, uploaded privately to the project team. Of course, we hope that some users will be feeling brave enough to share their recordings publicly.

The Visualising Voice project home page should be appearing online over the next few weeks, when we’ll share a bit more about the project. For now, watch this space…!

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s