Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Science

Scientists Re-Create Voice of 3,000-Year-Old Mummy (apnews.com) 39

Posted by BeauHD from the ancient-human-voices dept.
Longtime Slashdot reader vm writes: You don't have to wait until next Halloween to get creeped out. Using 3D printing, medical scanners, and an electronic larynx, researchers have recreated the voice of a 3,000-year-old Egyptian mummy. The tongue has deteriorated over three millennia and all they have so far is a vowel sound but it's a pretty clever way to raise the dead with science. "The researchers then synthesized Nesyamun's voice by 3D printing a model of his airway and connecting it to an electronic larynx, an artificial voice box that provides a noise source," reports Science Magazine. "Based on writings on Nesyamun's coffin and the objects he was buried with, researchers know that he was an Egyptian priest and scribe who likely sang and spoke to the gods as part of his ritual duties. His coffin inscriptions include a wish to 'see and address the gods as he had in his working life.'"

The findings have been reported in the journal Scientific Reports.
This discussion has been archived. No new comments can be posted.

Scientists Re-Create Voice of 3,000-Year-Old Mummy More

Scientists Re-Create Voice of 3,000-Year-Old Mummy

Comments Filter:

  • At last! (Score:5, Funny)

    by Volatile_Memory ( 140227 ) on Friday January 24, 2020 @09:09AM (#59651312)

    I have heard a 3,000 year old mummy say "Meh." Thanks, science!

  • Lol. Article links to audio file with three grunt like noises. Could be a cow, a kazoo, or ...a 3000 year old mummy's recreated voice. I'm going with cow.

    • Re:Very Underwhelming (Score:4, Interesting)

      by Xest ( 935314 ) on Friday January 24, 2020 @09:27AM (#59651376)

      Yeah, this was on the radio in the car on the way home yesterday, I was listening to the sounds and then the justification. The interviewer to be fair was asking legitimate questions like, how do you know he really sounded like this given none of the flesh is intact? They eventually admitted that the sound they've produced is the sound he'd make now, in his coffin, decomposed, if he could make sounds, so may share no resemblance with the sound the actual person would've made when he had an actual flesh tongue.

      As such, it all felt rather fucking pointless. It's certainly not what they're trying to sell it as; a recreation of his actual voice, merely sounds that they've managed to emanate from a reconstruction of his rotten carcass. It's like blowing over the top of a bottle to make that noise you can make with a bottle and claiming you've recreated the voice of the bottle, were it to be a living entity.

      • Re: (Score:2)

        by Nidi62 ( 1525137 )

        Yeah, this was on the radio in the car on the way home yesterday, I was listening to the sounds and then the justification. The interviewer to be fair was asking legitimate questions like, how do you know he really sounded like this given none of the flesh is intact? They eventually admitted that the sound they've produced is the sound he'd make now, in his coffin, decomposed, if he could make sounds, so may share no resemblance with the sound the actual person would've made when he had an actual flesh tongue.

        My understanding is that they took the measurements of his mouth/vocal tract and used what would be the "average" tongue for those sizes to create the sound. So while it is likely still off from his actual voice, it should be relatively close.

        Also, since the cadence, words, and pronunciations of religious chants are known, they think they could theoretically take the mock up and run it through a simulation, roughly reproducing the chants as he would probably have spoken them.

        • So not Gene Simmons then...

        • Re: (Score:2)

          by Xest ( 935314 )

          "My understanding is that they took the measurements of his mouth/vocal tract and used what would be the "average" tongue for those sizes to create the sound. So while it is likely still off from his actual voice, it should be relatively close."

          I'm not sure that's the case, certainly it contradicts what the researcher was saying yesterday. I also can't see anything relating to that in the study itself, though they do say the soft palate at least was estimated (because that was partially preserved):

          https://w [nature.com]

          • Re: (Score:3)

            by Nidi62 ( 1525137 )

            My info came from this article:
            https://www.cnn.com/2020/01/23... [cnn.com]

            Regarding the tongue, this is what it says:

            According to Howard, the team's Egyptian scholars said the phonetics and the music of the songs were known, so "in principle we could make him make different sounds and we could start to reproduce bits of what he actually sang." To do that, Howard said he would use computer software to build up the tongue based on the average for a vocal tract of that size.
            "Give him a tongue that's reasonable then we could move the whole of the vocal tract around using knowledge of speech production. It's feasible, although we can't do it easily at the moment."

            So they didn't actually recreate the tongue yet, but that would be required to produce the singing/chanting.

            It doesn't mention much about how the phonetics of the chants are known (always wondered how they figured out the characteristics of dead languages, just extrapolate from similar, known languages?). It would be interesting to see this used on well preserved remains where they can get all/most of

            • "always wondered how they figured out the characteristics of dead languages, just extrapolate from similar, known languages?" More or less. We also know (since this kind of extrapolation has been done for well over a century) the kinds of things that tend to happen to sounds over time--pretty much the same things we see in existing languages: consonants tend to get palatalized before front vowels, for example. So (simplifying a bit) Proto-Indo-European k turned into a sound like the 'ch' in 'chat' when i

    • Hmm, was pretty clear to me: "Those dang aliens are just sitting around drinking qahwa and are way behind schedule. Never going with the low bid again!"

  • This is the present state of science? I guess we're done now.

  • The first words were "Clean your room!"

  • Have they even done this on scans of a real person to see if it even works?

    • Re:Have they done this on an actual person to veri (Score:5, Informative)

      by Nidi62 ( 1525137 ) on Friday January 24, 2020 @10:07AM (#59651526)

      Have they even done this on scans of a real person to see if it even works?

      From an unlinked CNN article:

      Howard has already reproduced the vocal tracts of living people, including his own, using this same method and found that the sounds produced were very realistic. However, this is the first time the technique has been applied to human remains.....The project came about by chance when Howard discussed his work reproducing the vocal tracts of living humans with a former co-worker, and co-author of this paper, John Schofield from the University of York.
      "He looked at me and said: 'I'm an archaeologist. Could this be used for remains?'"

      So yes, they have.

      • Re: (Score:1)

        by HunterX ( 28292 )

        "He looked at me and said: 'I'm an archaeologist. Could this be used for remains?'"

        "'Cause I watched Jurassic Park 3 yesterday and seeing Alan Grant running around with a raptor kazoo was pretty damn awesome."

  • Ancient Egyptian (Score:3)

    by l0n3s0m3phr34k ( 2613107 ) on Friday January 24, 2020 @10:06AM (#59651520)
    From everything I've read, we actually have little idea exactly how ancient Egyptian would be vocalized. Like ancient Hebrew, it has no symbols for vowels. Once we can sort that out, the next obvious step is to have our robot start vocalizing rituals from The Book of the Dead. I think it would be brilliant to summon Ma'at; especially if Ma'at was summoned by a robot and therefor under the robot's bidding. There is a scenario that no anti-AI proponent ever though of: destruction of Humanity by AI via vengeful Egyptian gods!

    I, for one, welcome our new robotic summoned Ancient Egyptian God rulers!

    • Which would be understandable. I mean, what's your usual reaction when you notice it's a robocall?

    • Ancient Hebrew, like in the bible, has very extensive vowel symbols and punctuation marks in general, and even cantillation marks. In any case, it also has vowel letters that can be omitted and replaced with marks. The problem with "modern Hebrew" is that writing Hebrew with punctuation marks is far too slow, cumbersome and not practical, especially in the digital age, so people have to go by memory of what the punctuation of the word is supposed to be, and that can often be problematic.

      • Actually, those vowel marks were added in the middle ages. Ancient Hebrew (and other Semitic languages written in alphabetic scripts) lacked all those marks, and wrote only the consonants. The Greeks borrowed that alphabet, but invented symbols for vowels (or re-used symbols that represented consonants in Semitic, like aleph, for vowels: alpha).

        So the original poster is right: ancient Hebrew had no indication of vowels. And the medieval added vowel marks were the best guess about what the vowels were, ba

        • I know that it was added around that time, but actually didn't know that the middle ages are not considered "ancient times". In any case, Hebrew is not like ancient Egyptian, which never had scriptures that were widely passed to relatively "common people" and maintained for many generations by many and many types of religious people who dedicated their lives mainly to that. Egyptian scriptures and written language were always restricted to very few elite, so "ancient Egyptian" is as dead as a language can b
    • Apparently that's what Coptic, when fed to smart diachronic linguists, is for.
  • ...My voice is my passport. Verify Me. Hmmm...what's next, spoof voice auth?

    • Re: (Score:1)

      by skrugen ( 229044 )

      Exactly! The world isn't run by weapons anymore, or energy, or money, it's run by little ones and zeroes, little bits of data. It's all just electrons.

  • Do you want necromancers?? (Score:3)

    by xleeko ( 551231 ) on Friday January 24, 2020 @11:13AM (#59651804)

    ... because this is how you get necromancers. Honestly, don't these folks ever watch any movies? I think the film record speaks pretty clearly that this sort of thing is a BAD IDEA.

    "Great, now that we have his voice, what should we make him say?"

    "I don't know, how about something from this ancient book, bound in human skin and inked in blood?"

    "Great idea!!"

  • Sounds like the opening of a really bad horror movie...something staring Brendan Frasier!

  • Truly, this is relevant to the OP and NOT A PRON SITE despite the (I have to admit) extremely porn-sounding url.

    https://dood.al/pinktrombone/ [dood.al]

    It's a synthetic mouth you can maniuplate in realtime (still not a porn site) to hear the sounds produced from varying geometries.
    Kinda cool, although people will wonder WTF you're doing. If you tell them "oh I'm just playing with Pink Trombone" they will assume that is a porn site.

  • I thought he sounded like this: https://www.youtube.com/watch?... [youtube.com]

Slashdot Top Deals

He has not acquired a fortune; the fortune has acquired him. -- Bion

Close