I've written before about the frustration of losing expressive clarity when I lose my voice, so I am intrigued by Graham Pullin's work on expressive synthetic speech: Six Speaking Chairs (w/ Andrew Cook) and Speech Hedge (w/Ryan McLeod).
My voice loss is occasional and intermittent (as are its frustrations), but for people who can only speak through some type of augmentative or alternative communication (famously, Stephen Hawking and Roger Ebert), a significant challenge is the flatness of tone offered by most synthetic speech devices. These tools may offer slight changes in intonation for statements, questions, or exclamations but they don't offer the expressive range of the of the human voice.
As a quick example, consider the expressive impact of "These pretzels are making me thirsty" as delivered by a pretty decent text-to-speech tool and by the cast of Seinfeld. The impact of the statement, obviously, is not what is said, but how it is said.
(this example was generated with Loquendo/ Nuance.
Text-to-speech technology is constantly improving, but its current limitations in terms of expressive range are clear. We communicate information with our voices, but we also communicate excitement, love, frustration, doubt, fear, boredom--and on and on.
Six Speaking Chairs is both art installation and text-to-speech prototype. The chairs invite you to sit, to converse, to explore the parameters of expressive speech. Speech Hedge offers an opportunity to customize expression and tone, to give voice texture and subtlety. Both of these projects offer a space to explore the components of the expressive voice, to tease out what it is exactly that shapes and constrains the emotion we project with our voices.
Speech Hedge from Ryan McLeod on Vimeo.
Wednesday, January 23, 2013
Tuesday, November 6, 2012
Presidential Voices
Happy Election Day!
Here's a roundup of recordings, readings, and soundboards on presidential voices. Wouldn't someone you love like to get a call from Obama today?
- MSU Vincent Voice Library: A collection of voice recordings from U.S. presidents beginning with Benjamin Harrison recorded on an Edison wax cylinder.
- Greg Goodale on Presidential Voices, a great interview with WGBH. Romney sounds like the Fonz. ha!
- Presidential Voices, a book by Alan Metcalf.
- A collection of "Mittisms": My goodness gracious!
- Romney soundboard (from HuffPo)
- Presidential speech app: Get Obama (not sure if they've updated with a Romney version) to say whatever you want!
Wednesday, August 1, 2012
The Human Voice
A charming bit on the impact of the human voice from Studs Terkel and StoryCorps
Note: The captions are my first attempt at using Amara (formerly Universal Subtitles), an open-source captioning platform. Their captioning tools are very easy to use!
Thursday, May 3, 2012
Voices of the Dead
"We have already pointed out the startling possibility of the voices of the dead being reheard through this device." --"The Talking Phonograph," Scientific American, 1877
That recording devices could capture the voices of the dead was both a metaphorical and a literal proposition. With the invention of the phonograph, voices of the living could be captured and stored, rendering the speaker in some sense immortal. But there was speculation as well about whether the phonograph might be able to channel the voices of the dead or capture some essence of the living. Much of this speculation sprung from the spiritualism craze of the era, the phonograph just another medium (along with the ouija board and the telegraph) for the spirits to make contact.
There is even speculation that Edison was working on a "telephone" to contact the dead. He spoke of it in interviews, often glibly, but no specific records of the supposed invention's development exist. Regardless, Edison believed that if people were going to explore the afterlife, it should be through science, not speculation and show. Edison conceived of materiality as electricity, all of us composed of units of memory and experience, units that disperse rather than disappear upon death.1 It was unclear where these units might go, but perhaps there could be a machine to catch them. Somewhere between metaphor and matter, we might continue existing. Through a phonograph. A voicemail.
We've all had one of those voicemails, I suspect. The one we play and replay, saving it again and again. A person lost to us and suddenly reanimated through a simple voice. A ghostly presence, but more than a ghost. The voice not just a reminder of the person, but somehow the person made whole and present. A father is suing his mobile phone provider2 because a voicemail from his teenage daughter was deleted in the process of providing a trial service upgrade. His daughter died last year and this was his message, this was his connection to her. The case, of course, introduces many questions about who owns data and the overlapping of our private and public selves. While I can't speak to the implications of the case, I do understand the father's motives.
His daughter's recorded voice was more than just a token remembrance. This was his telephone to the dead.
1Rothman, A.D. "Mr. Edison's 'Life Units.'" The New York Times. 23 June 1921
2 You can read more about the case here.
That recording devices could capture the voices of the dead was both a metaphorical and a literal proposition. With the invention of the phonograph, voices of the living could be captured and stored, rendering the speaker in some sense immortal. But there was speculation as well about whether the phonograph might be able to channel the voices of the dead or capture some essence of the living. Much of this speculation sprung from the spiritualism craze of the era, the phonograph just another medium (along with the ouija board and the telegraph) for the spirits to make contact.
There is even speculation that Edison was working on a "telephone" to contact the dead. He spoke of it in interviews, often glibly, but no specific records of the supposed invention's development exist. Regardless, Edison believed that if people were going to explore the afterlife, it should be through science, not speculation and show. Edison conceived of materiality as electricity, all of us composed of units of memory and experience, units that disperse rather than disappear upon death.1 It was unclear where these units might go, but perhaps there could be a machine to catch them. Somewhere between metaphor and matter, we might continue existing. Through a phonograph. A voicemail.
We've all had one of those voicemails, I suspect. The one we play and replay, saving it again and again. A person lost to us and suddenly reanimated through a simple voice. A ghostly presence, but more than a ghost. The voice not just a reminder of the person, but somehow the person made whole and present. A father is suing his mobile phone provider2 because a voicemail from his teenage daughter was deleted in the process of providing a trial service upgrade. His daughter died last year and this was his message, this was his connection to her. The case, of course, introduces many questions about who owns data and the overlapping of our private and public selves. While I can't speak to the implications of the case, I do understand the father's motives.
His daughter's recorded voice was more than just a token remembrance. This was his telephone to the dead.
1Rothman, A.D. "Mr. Edison's 'Life Units.'" The New York Times. 23 June 1921
2 You can read more about the case here.
Labels:
history,
voice and relationships,
voice technology
Thursday, April 12, 2012
Monday, January 30, 2012
Wednesday, November 9, 2011
mask-bot: another talking head
Researchers in Germany and Japan (see Institute for Cognitive Systems) have collaborated to create this slightly creepy and cinematic talking head. The facial image is projected onto a mask to create the 3D likeness. Mask-bot doesn't respond through artificial intelligence, but through text-to-speech, voicing words typed into a keyboard.
Mask-bot is quite like Joseph Faber's Talking Machine, which I've written about previously. It's fascinating to see such similar results culled from technologies that are more than a century apart.
Labels:
history,
synthesized voices,
text-to-speech,
voice technology
Subscribe to:
Posts (Atom)