share


RETRIEVALS

 

About ‘Retrievals’:

 

I like to hear writing that is made out loud. Words vibrate in the air and you forget them, but you can feel them on your skin. I don’t call what I make ‘radio plays’. I just call them ‘audio pieces’. I like to keep it all as open as possible.

 

‘Retrievals’ is an audio piece, made using an online automated voice generator. There are many sites that offer the use of text-to-voice technology (Vocograb, Voxmark, NaturalReader). These websites can manufacture hundreds of different voices – men, women, children, the elderly – from across many different languages and dialects. They offer voices that sound sad, or whisper intimately in your ear. Some sites are free. Many make you pay for a particular voice.

 

Automated voices are produced for specific practical uses. They help the visually impaired, or those who have difficulty reading. They inform you where your train is going. They ask whether you want to pay with cash or card. They are calm and well-mannered. They are nearly always women. We do not listen to them, only overhear what they have to say. People who really listen in on them are often disturbed or put off, and programme their self-service checkout to ‘silent’ when they can.

 

Automated voices do not sound uncanny or robotic to me. They sound spectral and angelic. Each is a voice that once belonged to someone, each a literal remnant of recordings made by a voice actor, who provided all the phonemes, phrases and speech-parts, which are put together later. ‘Retrievals’ was made using a character called ‘Will (Sad)’, from acapela-box.com. The website contains no information concerning the real human being who was paid to perform the words for ‘Will (Sad)’. Any chance beauty of accent or inflection this voice might still possess remains only as the echo of something once heard, then lost, now forever misremembered. If automated voices sound ‘futuristic’ then it’s a backwards kind of future. They are forecasts of what has already been said.

 

I never keep what I’ve written for audio pieces; in this respect, voice-generator websites are ideal. You can type what you want into the little text boxes, listen back to how it sounds, edit accordingly, and delete it all once the recording has been made. I erase every last written word I’ve produced for these pieces. I love to do this. Deleting scripts is like the first time being kissed.


ABOUT THE CONTRIBUTOR

received his MFA from Washington University in St Louis, where he was Junior Writer in Residence (Fiction) 2015-16.

READ NEXT

feature

September 2012

Negation: A Response to Lars Iyer's 'Nude in Your Hot Tub'

Scott Esposito

feature

September 2012

I do not know whether I have anything to say, I know that I am saying nothing; I do...

Interview

June 2012

Interview with Malcolm McNeill

Patrick Langley

Interview

June 2012

I first met Malcolm McNeill in 2007. He was in London to do some printing for an exhibition, and he showed...

feature

June 2014

Hoarseness: A Legend of Contemporary Cairo

Youssef Rakha

feature

June 2014

U. Mubarak It kind of grows out of traffic. The staccato hiss of an exhaust pipe begins to sound like...