Welcome to the Simulation: Synthetic Speech

[elfismiles]

From 1999 to 2017 ... The rise and rise of synthetic speech.

synthetic speech
noun
1.
computer-generated audio output that resembles human speech, produced by an electronic synthesizer operated by means of a keyboard.

http://www.dictionary.com/browse/synthetic-speech

https://en.wikipedia.org/wiki/Speech_synthesis

I'll never forget, having read about the below 1999 demonstration, then being in a conference room filled with librarians in 2004 where some audio was played of a British sounding woman, and the tremendous communal gasp as the lecturer stated, "what you just heard was a completely synthetic voice."

#VoCo. Adobe MAX 2016 (Sneak Peeks) | Adobe Creative Cloud

https://www.youtube.com/watch?v=I3l4XLZ59iw

Published on Nov 4, 2016
#VoCo allows you to change words in a voiceover simply by typing new words. Presented live during the Adobe MAX 2016 Sneak Peeks, co-hosted by Jordan Peele. Learn more about this year's Sneak Peeks here: http://adobe.ly/2ffyder

LET’S GET EXPERIMENTAL: BEHIND THE ADOBE MAX SNEAKS
POSTED BY ADOBE CONVERSATIONS TEAM ON NOVEMBER 4, 2016

Adobe MAX Sneaks 2016
Does this sound familiar? You send your client a completed video project and they ask you to make a last minute change to the voiceover…but the voiceover artist is already on a plane to Hawaii. Well, thanks to the technology behind “Photoshopping Voiceovers,” this soon may no longer be an issue.

“Photoshopping Voiceovers,” or what we affectionately refer to as #VoCo, was one of 11 experimental technologies demoed at Adobe MAX 2016. The MAX Sneaks session invites our engineers out of the lab and onto the stage to show off what they’ve been working on. Co-hosted by television personality and comedian Jordan Peele and Adobe’s Community Engagement Manager Kim Chambers, this year’s sneaks had us alternating between laughing and gasping in awe.

Fortunately, you can enjoy the show even if you weren’t able to join us in San Diego. We’ve embedded videos of each of the sneaks below.

While these technologies are not yet part of Creative Cloud, many Sneaks from previous years have later been incorporated into our products. As always, we’d love your feedback. In fact, we’ve given each of the demos its own #hashtag for this very purpose. You’re welcome.

#VoCo

When recording voiceovers, dialogue, and narration, wouldn’t you love the option to edit or insert a few words without the hassle of recreating the recording environment or bringing the voiceover artist in for another session? #VoCo allows you to change words in a voiceover simply by typing new words. Have to hear it to believe it? Check out a live demo using a recording of co-host Jordan Peele’s voice.

https://blogs.adobe.com/conversations/2 ... neaks.html

What is Adobe Project VoCo? - Manipulate Voice | Adobe Max 2016

https://www.youtube.com/watch?v=M7hq0rEmxxQ

Adobe VoCo Review

https://www.youtube.com/watch?v=wvmCSOH4Okk

A Photoshop for voices. Listen to how real the results are.
https://www.facebook.com/techinasia/vid ... 054692105/

Adobe Voco 'Photoshop-for-voice' causes concern - BBC News
http://www.bbc.com/news/technology-37899902
Nov 7, 2016 - A new application that promises to be the "Photoshop of speech" is raising ethical and security concerns. Adobe unveiled Project Voco last ...

Adobe Voco - Should We Be Afraid? — Pro Tools Expert
http://www.pro-tools-expert.com/home-pa ... -be-afraid
Nov 19, 2016 - However, as Alan Sallabank highlighted in this story about Adobe Voco, which includes an Adobe demo that appears to be genuinely breaking ...


FLASHBACKs...

When Seeing and Hearing Isn't Believing

http://www.washingtonpost.com/wp-srv/national/dotmil/arkin020199.htm

"Once you can take any kind of information and reduce it into ones and zeros, you can do some pretty interesting things," says Daniel T. Kuehl, chairman of the Information Operations department of the National Defense University in Washington, the military's school for information warfare...

Pentagon planners started to discuss digital morphing after Iraq's invasion of Kuwait in 1990. Covert operators kicked around the idea of creating a computer-faked videotape of Saddam Hussein crying or showing other such manly weaknesses, or in some sexually compromising situation. The nascent plan was for the tapes to be flooded into Iraq and the Arab world.

<snip>

I stumbled upon this very interesting ad for a product that you can buy on the internet. The product supposedly can be used, not only to change your voice from male to female and vice versa, but even to provide a false caller id.


http://www.brickhousesecurity.com/spoof ... anger.html

And then I came across this article from the Washington Post, published in 1999:

When Seeing and Hearing Isn't Believing

Bill Arkin

By William M. Arkin
Special to washingtonpost.com
Monday, Feb. 1, 1999

"Gentlemen! We have called you together to inform you that we are going to overthrow the United States government." So begins a statement being delivered by Gen. Carl W. Steiner, former Commander-in-chief, U.S. Special Operations Command.

At least the voice sounds amazingly like him.

But it is not Steiner. It is the result of voice "morphing" technology developed at the Los Alamos National Laboratory in New Mexico.

By taking just a 10-minute digital recording of Steiner's voice, scientist George Papcun is able, in near real time, to clone speech patterns and develop an accurate facsimile. Steiner was so impressed, he asked for a copy of the tape.

Steiner was hardly the first or last victim to be spoofed by Papcun's team members. To refine their method, they took various high quality recordings of generals and experimented with creating fake statements. One of the most memorable is Colin Powell stating "I am being treated well by my captors."

"They chose to have him say something he would never otherwise have said," chuckled one of Papcun's colleagues.

A Box of Chocolates is Like War

Most Americans were introduced to the tricks of the digital age in the movie Forrest Gump, when the character played by Tom Hanks appeared to shake hands with President Kennedy.

For Hollywood, it is special effects. For covert operators in the U.S. military and intelligence agencies, it is a weapon of the future.

"Once you can take any kind of information and reduce it into ones and zeros, you can do some pretty interesting things," says Daniel T. Kuehl, chairman of the Information Operations department of the National Defense University in Washington, the military's school for information warfare.

PSYOPS seeks to exploit human vulnerabilities in enemy governments, militaries and populations.

Digital morphing — voice, video, and photo — has come of age, available for use in psychological operations. PSYOPS, as the military calls it, seek to exploit human vulnerabilities in enemy governments, militaries and populations to pursue national and battlefield objectives.

To some, PSYOPS is a backwater military discipline of leaflet dropping and radio propaganda. To a growing group of information war technologists, it is the nexus of fantasy and reality. Being able to manufacture convincing audio or video, they say, might be the difference in a successful military operation or coup.

Allah on the Holodeck

Pentagon planners started to discuss digital morphing after Iraq's invasion of Kuwait in 1990. Covert operators kicked around the idea of creating a computer-faked videotape of Saddam Hussein crying or showing other such manly weaknesses, or in some sexually compromising situation. The nascent plan was for the tapes to be flooded into Iraq and the Arab world.

The tape war never proceeded, killed, participants say, by bureaucratic fights over jurisdiction, skepticism over the technology, and concerns raised by Arab coalition partners.

What if the U.S. projected a holographic image of Allah floating over Baghdad?

But the "strategic" PSYOPS scheming didn't die. What if the U.S. projected a holographic image of Allah floating over Baghdad urging the Iraqi people and Army to rise up against Saddam, a senior Air Force officer asked in 1990?

According to a military physicist given the task of looking into the hologram idea, the feasibility had been established of projecting large, three-dimensional objects that appeared to float in the air.

But doing so over the skies of Iraq? To project such a hologram over Baghdad on the order of several hundred feet, they calculated, would take a mirror more than a mile square in space, as well as huge projectors and power sources.

And besides, investigators came back, what does Allah look like?

The Gulf War hologram story might be dismissed were it not the case that washingtonpost.com has learned that a super secret program was established in 1994 to pursue the very technology for PSYOPS application. The "Holographic Projector" is described in a classified Air Force document as a system to "project information power from space ... for special operations deception missions."

War is Like a Box of Chocolates

Voice-morphing? Fake video? Holographic projection? They sound more like Mission Impossible and Star Trek gimmicks than weapons. Yet for each, there are corresponding and growing research efforts as the technologies improve and offensive information warfare expands.

Whereas early voice morphing required cutting and pasting speech to put letters or words together to make a composite, Papcun's software developed at Los Alamos can far more accurately replicate the way one actually speaks. Eliminated are the robotic intonations.

The irony is that after Papcun finished his speech cloning research, there were no takers in the military. Luckily for him, Hollywood is interested: The promise of creating a virtual Clark Gable is mightier than the sword. (Well, we all know the Mossad has no access to "Hollywood"! -- Alice)

Video and photo manipulation has already raised profound questions of authenticity for the journalistic world. With audio joining the mix, it is not only journalists but also privacy advocates and the conspiracy-minded who will no doubt ponder the worrisome mischief that lurks in the not too distant future.

"We already know that seeing isn't necessarily believing," says Dan Kuehl, "now I guess hearing isn't either."

William M. Arkin, author of "The U.S. Military Online," is a leading expert on national security and the Internet. He lectures and writes on nuclear weapons, military matters and information warfare. An Army intelligence analyst from 1974-1978, Arkin currently consults for Washingtonpost.Newsweek Interactive, MSNBC and the Natural Resources Defense Council.

Arkin can be reached for comment at william_arkin@washingtonpost.com.

© Copyright 1999 The Washington Post Company

http://www.washingtonpost.com/wp-srv/na ... 020199.htm


Hmmm. Eerily accurate voice morphing, based on a 10-second recording of the target's voice... But where on earth could such a recording be obtained?

From an Amy Goodman interview with James Bamford, author of The Shadow Factory: The Ultra-Secret NSA from 9/11 to the Eavesdropping on America:

JAMES BAMFORD: Yeah. There’s two major—or not major, they’re small companies, but they service the two major telecom companies. This company, Narus, which was founded in Israel and has large Israel connections, does the—basically the tapping of the communications on AT&T. And Verizon chose another company, ironically also founded in Israel and largely controlled by and developed by people in Israel called Verint.

So these two companies specialize in what’s known as mass surveillance. Their literature—I read this literature from Verint, for example—is supposed to only go to intelligence agencies and so forth, and it says, “We specialize in mass surveillance,” and that’s what they do. They put these mass surveillance equipment in these facilities. ...

quoted here

The Schiphol Airport once again comes to the fore...


<snip>

[liminalOyster]

Good thing this didn't exist in the early years after 9/11 or some clown might've called those very really totally very authentic OBL recordings into question. Soon enough I hope we'll finally gain the technology to remotely control car brakes, allowing us to make it appear that man-sticking investigative journalists are simply big loser paranoid drug addicts. Dare to dream.

[smoking since 1879]

“Believe none of what you hear, and only half of what you see.”
― Benjamin Franklin

[EDIT] ... or Edgar Allan Poe or Mickey Mouse ... depends where you read it

[tron]

Good thing this didn't exist in the early years after 9/11 or some clown might've called those very really totally very authentic OBL recordings into question. Soon enough I hope we'll finally gain the technology to remotely control car brakes, allowing us to make it appear that man-sticking investigative journalists are simply big loser paranoid drug addicts. Dare to dream.

some bright spark has integrated blue tooth in the car brake system.

[82_28]

Can I get a resounding Philp Kindred Dick is spinning in his grave? Yow. Faking anything is possible now.

Actually maybe PKD would have loved this.

Also can you take old audio and change the past? I would suppose so. Wow. This scares me.

[82_28]

Oh. It reminds me of this time myself and some friends as kids had crushes on girls at our church's youth group. We would call the girls and ask them random questions trying to get them to say something that we would edit in ANALOG cassette tape -- with physical splicing. I don't know why we did it but we did. We didn't ever let it out, we just did it to see if we could. I think. It was just a fun adolescent project, as it were. I think we did it with one of those dual cassette boom boxes.

[liminalOyster]

Good thing this didn't exist in the early years after 9/11 or some clown might've called those very really totally very authentic OBL recordings into question. Soon enough I hope we'll finally gain the technology to remotely control car brakes, allowing us to make it appear that man-sticking investigative journalists are simply big loser paranoid drug addicts. Dare to dream.

some bright spark has integrated blue tooth in the car brake system.

Well that's sensible. Some pacemakers have had bluetooth data download (at least) for a few years too.

[IanEye]

dave: i used to have a synton syntovox 221 vocoder, but I flogged it to speedy j in the 90's, and I wish I still had it. I got it & bunch of other klob as payment for some secret crypto work when I was at bt research labs, which I can't go into detail on coz I signed the official secrets act in 1981 when I started as an apprentice. can u tell me a bit about how u used your pair of em? (I rem when u wrote to me telling me u had 2, and u could go realtime syntovox stereo!

rich: exactly, +an excuse to buy another, they are slightly different as well.
Apart from the patch ability which is so nice on it with pins [like ems5000] it has the computer interface, which always intrigued me, if synton did anything with that or anyone else, or you?

dave: well not at home, but we had some interesting stuff at the labs there on martlesham heath to interface to all the speech recognition / synthesis klob. but i can;t really talk bout the cloak & dagger stuff except to say tht its now been officially declassified in 2003 about MI5, MI6, GCHQ et al at least being on site. see this link & search for martlesham :

http://www.informationclearinghouse.info/article4464.htm

MI6 Telephone intercepts were for some time handled at the London Station or VBR, by a group of specialists and linguists known as UKZ and operating with a team of specially cleared BT engineers known as the OND. Metropolitan Police Interception and Special Services Centre was situated at 113 Grove Park, Camberwell, London SE5 and served as a joint MI5/MI6/MPSB/C7/GC & CS unit. This had been in operation as 'Grove Park' since around 1919 and was still a covert listening site well into the 1980's. Some operations were transferred to Sandridge near St Albans in the late 1930's and that base was taken over by GCHQ in 1946. A fleet of detector vans was based there throughout the 1950's and 1960's. By 1970's had reverted to Home Office control and had became a Surveillance Research centre developing equipment for Gove Park and other users.

The secret R12 Department at the Martlesham Heath Telecommunications Complex near Ipswich is a major R & D source of surveillance technology and works closely with GCHQ/MI5/MI6/NCIS, while similar work was also carried out at the JSERL-Joint Services Electronics Research Laboratory at Baldock in Hertfordshire(former wartime GPO DF Station along with Burnham), where bugging equipment for use in Ulster and by both MI5 and MI6 was developed. MI5 interceptions are made via the BTID facilities on the 25th and 26th Floors of the Euston Tower and fed to the Transciption Unit at Thames House via a secure digital line. Since 1995 the transcribers have used the 'Marshbrook' computer system and updated versions which transcribe, analyse, log and file all interception communications. 2003 - Current Hi-Tech System controlled from the BT National Network Central Operations Unit, National Special HQ, Broggyntyn Hall, Oswestry, Shropshire, for many years the National Emergency Network Control Centre. (Microwave Network Link at ALBRIGHTON).

BT Worldwide Network Management Centre at Oswestry, Shropshire, was opened on 5 September 1990 at a cost of £4 million. The Centre monitored all of BT's System X exchanges (57 trunk and 373 local exchanges) and the company's three digital internal exchanges. Has overall control of interception of international calls. The new National Network Control Centre officially opened on September 12, 2002.The video wall measures 16.32m by 3.06m made up of 36 monitors. The giant screen is the same as used in modern digital cinemas with more than 800,000 individual mirrors behind the screen, a solid-state system that ensure razor sharp images. The control centre was developed for BT by CCC Network Systems using FreeVision technology: the wallboard was supplied by Synelec. The £10 million state-of-the-art building at Oswestry in Shropshire, gives engineers a helicopter view of the state of health of the UK's communications networks. They can view telephone call, data and broadband traffic at a glance, responding to incidents world-wide which could impact on quality of service not only to BT's customers but also to other UK operators and service providers and helps HMG and the Intelligence Community have both oversight and eventual control of the entire network. Has overall control of the interception of national calls.

Britain’s telephone network, System X, was designed with wiretapping capabilities built in and indeed all the digital exchanges built since the mid 1990’s have an intercept capability under the aegis of the European Telecommunications Standards Institute.

there's also a tv program that itv made called 'the buggist' which got a bit bout the bt labs MI5 et al connections :

http://www.itnsource.com/shotlist//ITVProgs/1989/04/04/Y05870075/?s=intv

aint seen the program, but they won;t b tellin me anything new anyway i'm not endorsing those sites, but they do note the presence of MI5 et al at BT Labs as it was then... anyway, it is weird to think that there's this whole area of electronic sound/noise research that never gets published or patented and all happens in mysterious little rooms, wicked & wonderful sounds that would make u freak! all these nutty top secret processes & obscure methods, exciting stuff but also bit weird coz i can never talk about it unless they declassify it... which is unlikely. funny tho, theres been a little bit about where vocoders came from in secret speech coding.

rich: bonkatudinal.

dave: e.g. http://boingboing.net/2010/10/06/vocoders.html

rich: wick will dive into dat

dave: my roland system 100m came to me the same way, for doing secret work for em, sort of little bonus, they used to use it for simulating dial tone / test tones in phone testing back in the 80's! nuts...

rich: ha, really funny weird, kind of remem u saying that prob after i necked half a bottle of absinth or summink

dave: yea, have some vague memory of drinking absynth with ya... ! i also think there's prob lot more speech synthesis/recog/vocoder stuff been made by the big "music manufacturers" that gets diverted to more secret uses. so, yea that interface on the back of the synton was specially designed for hooking up to pooter for analysis/resynthesis, its do-able ... cripes, i could talk about this all day, .. anyway u were saying about synton ...

rich: yeah i thought you prob could, ok save it for when i interview you


I mean I suppose the main thing to do there is recording the cv outs from the filters of analysed audio, manipuating the voltages with a seq then chuck in it back at the vocoder, did try this a few times with a doepfer setup, theres definitly plenty of scope their for future experiments, give me a clone please someone?
also check this out on the dutch speech synthesis tip, i got it from Steim about 15 years ago, VOSIM voice simulation sound synthesis,
iits based on the idea that by employing repeating tone-burst signals of variable pulse duration and variable delay, a sound output of high linguistic and musical expressive power can be obtained



.

[identity]

I am always surprised that the entertainment applications of this technology have not yet been deployed; obviously—inevitably—at some point (which I thought would have arrived long, long ago) it will be possible for music producers, and eventually even home users, using the appropriate software, to apply the timbral characteristics of any voice to any other, so that with a bit of fiddling about, you could hear Elvis singing Stairway to Heaven, or Nick Drake singing Billy Jean (!). I would have thought that we are now far beyond the point where this is easily doable with existing technologies. (And if it isn't the case yet, it will, without any question, be so in the not-too-distant future.)

[dada]

1982.

Since the Intellivision was based on a General Instruments chip set, it didn't take long for someone to realize that the General Instruments "Orator" speech-synthesis chip could be the basis for a nifty Intellivision add-on. ...the voices for the first voice game, Space Spartans, were recorded and digitized in New York, at General Instruments' voice lab. Ron Carlson and Patrick Jost supervised the sessions in New York and sent the data back to Ron Surratt in California, who loaded it into the Intellivoice prototype. But once switched on, all it could do was keep repeating "Auk yooo! Auk yooo!" which didn't go over well with the Mattel executives and marketing personnel. In several heated phone calls between Hawthorne and New York, Carlson blamed the problem on Surratt's software and Surratt blamed it on Carlson's hardware.

"I really didn't know what I was talking about," Surratt admits today, "but luckily it did turn out to be the hardware."

I really didn't know what I was talking about, but luckily it did turn out to be the hardware. Funny guys over there. I don't know how they got any work done.

"Hello Commander, computer reporting"
http://intellivisionlives.com/bluesky/voice/hello/HelloCom.wav

Intellivision sure has a sexy robot voice. auk yooo.

One way to insure that your voice can't be photoshopped and used against you is to talk like a grindcore singer, all the time. On the phone, ordering at the drive thru, at the supermarket. People might look at you strange. Just scream, "I'm keeping my vooooice... from being uuuuused.... against meeee!" Then they'll call the cops.

[DrEvil]

I am always surprised that the entertainment applications of this technology have not yet been deployed; obviously—inevitably—at some point (which I thought would have arrived long, long ago) it will be possible for music producers, and eventually even home users, using the appropriate software, to apply the timbral characteristics of any voice to any other, so that with a bit of fiddling about, you could hear Elvis singing Stairway to Heaven, or Nick Drake singing Billy Jean (!). I would have thought that we are now far beyond the point where this is easily doable with existing technologies. (And if it isn't the case yet, it will, without any question, be so in the not-too-distant future.)

I'm looking forward to the gaming applications myself. Pair up voice synthesizing with a game specific version of Siri and go discuss philosophy with the barman in Elder Scrolls 9 or get into arguments with random pedestrians in Grand Theft Auto 8 before shooting them.

We will also probably see celebrities licensing their synthesized voices for commercials and animated movies, anything where their faces don't appear (judging by the digital characters in Rogue One there's still some work to be done before virtual actors are a thing, at least cheaply enough to be feasible. Peter Cushing and Carrie Fisher both looked fake as hell, and that was with some of the world's best digital artists on the job), and then soon afterwards Mel Gibson will be prank calling people with custom made racist tirades, and then someone will impersonate someone important to do something really bad/magnificent/stupid and then it will become just another thing that people take for granted.