# Why does Stephen Hawking's voice sound so robotic?

**URL:** <https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937>\
**Category:** Factual Questions\
**Created:** [February 9, 2008, 5:33pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937 "2008-02-09T17:33:35Z")\
**Posts on this page:** 20\
**Page:** 1

<div class="post-metadata">

**Author:** ![aldiboronti](https://avatars.discourse-cdn.com/v4/letter/a/9fc348/32.png) [@aldiboronti](https://boards.straightdope.com/u/aldiboronti)\
**Post date:** [February 9, 2008, 5:33pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/1 "2008-02-09T17:33:35Z")

</div>

What’s the deal here? Surely synthesizers these days can come up with something a little more human sounding. Or is it just very old hardware which hasn’t been updated?

---

<div class="post-metadata">

**Author:** ![Busy\_Scissors](https://avatars.discourse-cdn.com/v4/letter/b/d78d45/32.png) [@Busy\_Scissors](https://boards.straightdope.com/u/Busy_Scissors)\
**Post date:** [February 9, 2008, 5:42pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/2 "2008-02-09T17:42:26Z")

</div>

[QUOTE=aldiboronti]  
What’s the deal here? Surely synthesizers these days can come up with something a little more human sounding. Or is it just very old hardware which hasn’t been updated?  
[/QUOTE]  
I believe he likes it, having gotten used to it for so long it’s like his own voice and is reluctant to change it. I thought I read this on his wiki page, but can’t see it the noo. Anyway, this is the explanation that I have heard.

---

<div class="post-metadata">

**Author:** ![Scarlett67](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/scarlett67/32/7387_2.png) [@Scarlett67](https://boards.straightdope.com/u/Scarlett67)\
**Post date:** [February 9, 2008, 6:01pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/3 "2008-02-09T18:01:00Z")

</div>

[QUOTE=The Cocky Watchman]  
I believe he likes it, having gotten used to it for so long it’s like his own voice and is reluctant to change it.  
[/QUOTE]

What I’ve read is that he’s unhappy that his voice has an American accent. Apparently there aren’t any with English accents, at least not that are up to snuff.

---

<div class="post-metadata">

**Author:** ![Squink](https://avatars.discourse-cdn.com/v4/letter/s/b5e925/32.png) [@Squink](https://boards.straightdope.com/u/Squink)\
**Post date:** [February 9, 2008, 6:08pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/4 "2008-02-09T18:08:08Z")

</div>

[QUOTE=Scarlett67]  
What I’ve read is that he’s unhappy that his voice has an American accent.  
[/QUOTE]  
80’s vintage voice synthesizers were doing well to produce intelligible speech at all, much less proper accenting.

---

<div class="post-metadata">

**Author:** ![SharkB8](https://avatars.discourse-cdn.com/v4/letter/s/977dab/32.png) [@SharkB8](https://boards.straightdope.com/u/SharkB8)\
**Post date:** [February 9, 2008, 6:10pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/5 "2008-02-09T18:10:12Z")

</div>

Some of the voices on a Mac have British Accents…

---

<div class="post-metadata">

**Author:** ![CookingWithGas](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/cookingwithgas/32/485_2.png) [@CookingWithGas](https://boards.straightdope.com/u/CookingWithGas)\
**Post date:** [February 9, 2008, 7:11pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/6 "2008-02-09T19:11:41Z")

</div>

[QUOTE=Scarlett67]  
What I’ve read is that he’s unhappy that his voice has an American accent. Apparently there aren’t any with English accents, at least not that are up to snuff.  
[/QUOTE]  
My GPS can speak English in American, British, Australian accents, plus about a dozen or so different languages. They are very understandable.

---

<div class="post-metadata">

**Author:** ![bouv](https://avatars.discourse-cdn.com/v4/letter/b/bc8723/32.png) [@bouv](https://boards.straightdope.com/u/bouv)\
**Post date:** [February 9, 2008, 8:48pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/7 "2008-02-09T20:48:01Z")

</div>

[QUOTE=CookingWithGas]  
My GPS can speak English in American, British, Australian accents, plus about a dozen or so different languages. They are very understandable.  
[/QUOTE]

Your GPS isn’t from the 80’s.

---

<div class="post-metadata">

**Author:** ![Indistinguishable](https://avatars.discourse-cdn.com/v4/letter/i/90ced4/32.png) [@Indistinguishable](https://boards.straightdope.com/u/Indistinguishable)\
**Post date:** [February 9, 2008, 9:55pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/8 "2008-02-09T21:55:04Z")

</div>

It’s not like British accents are intrinsically more complex to synthesize than American accents or anything, of course. But I guess all the initial work on such things was being done by Americans, and thus what was first available was stuff with American accents.

---

<div class="post-metadata">

**Author:** ![essell](https://avatars.discourse-cdn.com/v4/letter/e/34f0e0/32.png) [@essell](https://boards.straightdope.com/u/essell)\
**Post date:** [February 9, 2008, 10:35pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/9 "2008-02-09T22:35:29Z")

</div>

[QUOTE=CookingWithGas]  
My GPS can speak English in American, British, Australian accents, plus about a dozen or so different languages. They are very understandable.  
[/QUOTE]

I don’t know what GPS you have, but the ones I’ve seen all have recorded clips played in different combinations. It’s not voice synthesis.

It’s much easier as they have a lot less things they might need to say.

---

<div class="post-metadata">

**Author:** ![friedo](https://avatars.discourse-cdn.com/v4/letter/f/8edcca/32.png) [@friedo](https://boards.straightdope.com/u/friedo)\
**Post date:** [February 9, 2008, 10:40pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/10 "2008-02-09T22:40:44Z")

</div>

[QUOTE=essell]  
I don’t know what GPS you have, but the ones I’ve seen all have recorded clips played in different combinations. It’s not voice synthesis.

[/QUOTE]

Some of the newer ones use voice synthesizers (they’re a lot better than in the old Macintalk days). They are spiffy because they (try to) pronounce every road name and destination, which would be infeasible for pre-recorded systems.

---

<div class="post-metadata">

**Author:** ![pulykamell](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/pulykamell/32/3166_2.png) [@pulykamell](https://boards.straightdope.com/u/pulykamell)\
**Post date:** [February 9, 2008, 10:43pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/11 "2008-02-09T22:43:09Z")

</div>

[QUOTE=essell]  
I don’t know what GPS you have, but the ones I’ve seen all have recorded clips played in different combinations. It’s not voice synthesis.

It’s much easier as they have a lot less things they might need to say.  
[/QUOTE]

The one I use, the nüvi, has voice synthesis and sounds very natural indeed.

[Link.](http://egovsergo.com/2007/07/04/garmin-nuvi%C2%AE-360/)

> [@](#):
>
> The nüvi also boasts a fantastic array of voices driven by their Text-to-Speech (TTS) engine to read street names. Unlike previous Garmin products, all of the speech is synthetic (when a TTS voice is used). I prefer Virginie and Daniel. I must stop the casual reader here and explain why I think my ‘approval’ of this voice system is meaningful. I am a graduate student in linguistics and my thesis is based in phonetics and phonology. I have a very well-trained ear; for a voice synthesis system to impress me, it must be quite good. I’m impressed.

---

<div class="post-metadata">

**Author:** ![essell](https://avatars.discourse-cdn.com/v4/letter/e/34f0e0/32.png) [@essell](https://boards.straightdope.com/u/essell)\
**Post date:** [February 9, 2008, 10:50pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/12 "2008-02-09T22:50:35Z")

</div>

[QUOTE=friedo]  
Some of the newer ones use voice synthesizers (they’re a lot better than in the old Macintalk days). They are spiffy because they (try to) pronounce every road name and destination, which would be infeasible for pre-recorded systems.  
[/QUOTE]

[QUOTE=pulykamell]  
The one I use, the nüvi, has voice synthesis and sounds very natural indeed.  
[/QUOTE]

Oh. Nevermind then.

I guess Stephen Hawking doesn’t want to change his voice now. As clearly he could.

---

<div class="post-metadata">

**Author:** ![pulykamell](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/pulykamell/32/3166_2.png) [@pulykamell](https://boards.straightdope.com/u/pulykamell)\
**Post date:** [February 9, 2008, 10:51pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/13 "2008-02-09T22:51:17Z")

</div>

[Actually, looking up some TTS online, I found this example, which is rather remarkable.](http://www.oddcast.com/home/demos/tts/tts_example.php)

---

<div class="post-metadata">

**Author:** ![Indistinguishable](https://avatars.discourse-cdn.com/v4/letter/i/90ced4/32.png) [@Indistinguishable](https://boards.straightdope.com/u/Indistinguishable)\
**Post date:** [February 9, 2008, 10:56pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/14 "2008-02-09T22:56:24Z")

</div>

[QUOTE=pulykamell]  
[Actually, looking up some TTS online, I found this example, which is rather remarkable.](http://www.oddcast.com/home/demos/tts/tts_example.php)  
[/QUOTE]

Charles (UK) is awesome. I’m having him read my category theory textbooks, line by line.

---

<div class="post-metadata">

**Author:** ![Indistinguishable](https://avatars.discourse-cdn.com/v4/letter/i/90ced4/32.png) [@Indistinguishable](https://boards.straightdope.com/u/Indistinguishable)\
**Post date:** [February 9, 2008, 11:09pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/15 "2008-02-09T23:09:40Z")

</div>

Damn! Time limit… oh, how can I slog through these papers now? The grandeur is lost; I can’t bear to go back to that drab, dull world of before.

---

<div class="post-metadata">

**Author:** ![pulykamell](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/pulykamell/32/3166_2.png) [@pulykamell](https://boards.straightdope.com/u/pulykamell)\
**Post date:** [February 9, 2008, 11:18pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/16 "2008-02-09T23:18:52Z")

</div>

[QUOTE=Indistinguishable]  
Damn! Time limit… oh, how can I slog through these papers now? The grandeur is lost; I can’t bear to go back to that drab, dull world of before.  
[/QUOTE]

Oh, shoulda warned you. Same thing happened to me. I also tried it in Polish, and wow. Sounds great.

---

<div class="post-metadata">

**Author:** ![Indistinguishable](https://avatars.discourse-cdn.com/v4/letter/i/90ced4/32.png) [@Indistinguishable](https://boards.straightdope.com/u/Indistinguishable)\
**Post date:** [February 9, 2008, 11:20pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/17 "2008-02-09T23:20:33Z")

</div>

Going [here](http://vhost.oddcast.com/admin/sitepalV4.php?spec=1), clicking “Let’s start” and then going to TTS under button 4 seems to get around the time limit, at least so far.

I’m not getting any work done today…

---

<div class="post-metadata">

**Author:** ![minor7flat5](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/minor7flat5/32/258_2.png) [@minor7flat5](https://boards.straightdope.com/u/minor7flat5)\
**Post date:** [February 10, 2008, 2:30am UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/18 "2008-02-10T02:30:17Z")

</div>

Here’s some interesting info about Stephen Hawking’s machine:

Oram, Andy, and Greg Wilson, eds. Beautiful Code. Sebastopol, CA: O’Reilly, 2007. 483-484.

> [@](#):
>
> Professor Hawking, the eminent theoretical physicist, has ALS. This disease is “marked by gradual degeneration of the nerve cells in the central nervous system that control voluntary muscle movement. The disorder causes muscle weakness and atrophy throughout the body.” He writes and speaks using the software Equalizer, which he operates via a single button. **It uses an external box for text-to-speech, which is no longer manufactured. The source code for Equalizer has been lost.**  
> …  
> Professor Hawking **still continues to use the software Equalizer, which he has been familiar with for decades.**

(Emphasis mine)

The chapter, entitled “When a Button Is All That Connects You to the World” was written by Arun Mehta, a co-founder of Radiophony, a company that is working on an open-source product called [eLocutor](http://www.holisticit.com/eLocutor/elocutor.ppt)(Warning: PowerPoint), in hopes of replacing the aging setup that Professor Hawking uses.

Here’s the Radiophony page on [The Hawking Project](http://www.radiophony.com/html_files/hawking.html)

Here’s [Professor Stephen Hawking’s description of his experience with ALS](http://www.hawking.org.uk/text/disable/disable.html)

---

<div class="post-metadata">

**Author:** ![Harmonious\_Discord](https://avatars.discourse-cdn.com/v4/letter/h/74df32/32.png) [@Harmonious\_Discord](https://boards.straightdope.com/u/Harmonious_Discord)\
**Post date:** [February 10, 2008, 2:56pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/19 "2008-02-10T14:56:03Z")

</div>

One thing that has hampered the text to speech readers, is the copyright arguing going on. Often books were published and then a different company bought rights to sell it on cassette. The audio books sellers didn’t like that people used a reader on the text file books. All sorts of restrictions and stuff during this long fight put off many developers. I think we could have been further along than we are in this field.

I have a British male and female voice that I use occasionally for text to speech books.

---

<div class="post-metadata">

**Author:** ![mobo85](https://avatars.discourse-cdn.com/v4/letter/m/b19c9b/32.png) [@mobo85](https://boards.straightdope.com/u/mobo85)\
**Post date:** [February 10, 2008, 9:34pm UTC](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937/20 "2008-02-10T21:34:51Z")

</div>

[QUOTE=pulykamell]  
[Actually, looking up some TTS online, I found this example, which is rather remarkable.](http://www.oddcast.com/home/demos/tts/tts_example.php)  
[/QUOTE]

The voices appear to be from [AT&T Labs TTS](http://www.research.att.com/~ttsweb/tts/demo.php), which is one of the best (or at least the most “human-sounding”) TTS I’ve heard.

[Next page](https://boards.straightdope.com/t/why-does-stephen-hawkings-voice-sound-so-robotic/436937.md?page=2)
