Roger Ebert, “Remaking my Voice” (2011) [idea clutch]

What does it mean to “have a voice” in the age of the internet?

“These are my words but this is not my voice,” the robotic computer voice reads as Roger Ebert, seated next to his wife and two friends, looks out into the audience. Film’s most prominent critic quite literally lost his voice to throat cancer and this is the talk about his journey, both emotionally and technically, to finding it. 

Each time I watch it, it takes on new meaning for me.

1. A talk about intimacy

Ebert’s talk alternatingly delivered via Roger’s computer, his wife Chaz, and two friends reading on his words on his behalf. As these people close to him share his words, Ebert mimes; he points his two thumbs down when Chaz describes a failed surgery, he makes a tunnel with his hands around his face as Chaz explains how the mouth, tongue, and throat work together to make a voice. These gestures remind the viewer to look at him as they listen. “These words belong to me. I inhabit them.”

Several times throughout the talk, Chaz, Ebert’s wife, reaches over and clutches her husband’s arm. She tears up a few times. In one segment, Chaz reads “I look like The Phantom of the Opera” from Ebert’s script and then interjects with “no you don’t!” as if in true conversation with him. I tear up when she does this. I think it’s because they’ve found a unique and deep way to connect and converse in these impossible circumstances.  

About a year after Ebert’s death in 2013, Chaz appears on the podcast,Death Sex and Money, and describes this connection. “At some point, he asked me to be his voice. When he was very sick, it felt like we became one person. I didn’t feel the boundaries that you feel with two people. And I know those boundaries so well because when he got better and he got stronger, those boundaries were resurrected… In the periods when we became one... I actually did feel him in my soul when we became one person.” 

2. A talk about the benefits, and limitations, of technology

Without his voice, Ebert describes “feel(ing) distant from human mainstream.” Talking is such a foundational and intimate element of connecting. I hadn’t thought about it much before, but it’s so clear to me now that conveying words or ideas is only one small part of talking. The way you bend and manipulate and style the words with your voice reflects the texture and personality that is uniquely yours. 

It’s what’s so frustrating to Ebert, and so many others, who have to use voice dictation software to express themselves. Ebert, a man of means with unlimited connections, couldn’t find software that was better than the voice that came-out-of-the-box with any Mac (called Alex). He and Chaz found a company that was willing to take hundreds of hours of footage from Ebert’s television and radio days and stitch all the bits and pieces of syllables and intonation together into a functioning and personalized voice. Even that wasn’t great; the words didn’t bend or move as Ebert actually would have bent them. He opted to continue just using Alex.

The technology required to do this kind of thing is getting a lot better. For people like Ebert who don’t have a voice, this work is so meaningful. When I hear my Alexa or the voice on Google maps, I feel the distance Ebert talks about when he has to use his computer voice in place of his actual, human, voice. Talking to my friend shouldn’t feel like talking to my device.

I got interested in this idea and came across VocalID, a company created by Rupal Patel (voice scientist and professor) that builds customized voices for people who use synthetic speech. She was inspired to start this project when she witnessed a young girl and older man interacting at a conference with the exact same computer voice (which, by the way, is a man’s voice by default). “We wouldn’t dream of fitting a little girl with the prosthetic limb of a grown man so why then the same prosthetic voice ,” she said on TED Radio Hour.  VocalID matches the tone of her clients voices (even those with severe speech disorders) to a “surrogate talker,” who has donated her voice to the project. She blends them together to create a distinctive, differentiated, voice for each client. How beautiful is that?

Ebert closes his talk with a short, punchy joke, told via his computer voice. The audience laughs out loud. Ebert says his voice is “good enough” if the voice’s comedic timing is good enough to elicit a laugh. Still, I bet Ebert wished his computer voice it could do more. A lot more. 

3. (Most recently) A talk about what it means to “have a voice” in the age of The Internet

I have probably read hundreds of Ebert’s film reviews in my lifetime (I watch a lot of movies). The operative word here is “read.” The voice I have in my head as I read it never had anything to do with his actual voice. I made it up based on what I imagined him to be and sound like. My parents, on the other hand, probably know exactly what Ebert sounds like. He and his partner critiquer Gene Siskel hosted a very popular television show called Sneak Previews that debuted in the mid-70’s and ran for almost 20 years. 

My connection to Ebert’s “voice” has everything to do with his online voice; his written reviews, his blog posts, and his Twitter. It didn’t occur to me the extent to which I do this for many people that I follow on the internet. Ebert says in this TED talk “my ability to write on the internet has been a life saver.” Ebert had a better shot than anybody not to “need” an actual voice, and still, for obvious reasons, it is such a profound loss not to have one. 

The internet, and the information that spread through it, didn’t sound like much until recently. Outside of straightforward music, the content on the internet was full of words and pictures, but not much audio. That’s changing; audio is absolutely exploding. Spotify has very publicly made it a named priority. A16z’s (very comprehensive and interesting) report on the podcasting ecosystem states that a quarter of Americans are listening to podcasts weekly and 65% of them have started listening in the last 3 years. “Alexa” is becoming ubiquitous; at this time last year, Apple reported having sold 100 million devices. Audiobook sales have been up double digits for the past six years in a row. 

I caught the podcasting bug early. For me, it feels different to consume things through my ears than my eyes. There’s an intimacy to have someone’s voice in your ear, even a stranger. When it comes to editorial and ideas, audio feels a vehicle for deeper correspondence and connection. 

The role of audio and voice in other technology is less straightforward to me. My friend’s grandmother is getting pretty good at using her Alexa. She always  “thank you” after she plays the song or sets the timer or adds the item to her grocery list. I think she feels like it’s a person and she should be polite. I feel differently. To me, even though there is a voice there, she still feels like a computer. I don’t feel the urge to be particularly polite. Maybe that’s a good thing. Maybe that will change. 

More to buzz about:

  • My Personal Moonshot by Tyler Cowen is, by far, my favorite post on the topic of writing/having a voice on the internet

  • In 1975, Ebert became the first film critic to win the Pulitzer Prize for Criticism.

  • Ebert was one of the pioneers of writing on a computer and writing online. He describes buying one of first model 100 computers at Radio Shack and recommending to the salesperson that they hook the shop up to the internet.  

  • You can inquire about “donating” your voice to Rupal Patel’s VocalID project here