"DNA-based prediction of Nietzsche's voice"

« previous post | next post »

An interesting paper was recently brought to my attention: Flavia Montaggio, Patricia Montaggio, & Imp Kerr, "DNA-based prediction of Nitzsche's voice", Investigative Genetics, Spring 2015. The abstract is pretty good:

This paper presents a protocol for the accurate prediction of an individual’s voice based on genotype data, specifically from single nucleotide polymorphisms (SNPs). We collected trace amounts of cellular material (Touch DNA) from books that belonged to the philosopher Friedrich Nietzsche (1844-1900). DNA was extracted and amplified using DOP-PCR technique. Five different genomic DNAs were generated. Nietzsche’s genotype was singled out after comparison to genotype data from one living relative of the Nietzsche family. Nietzsche’s genotype data was analyzed using a DNA-based phenotyping assay, termed VoiceRator, that incorporates the 24 most informative voice SNPs based on their association with genes related to the phenotypic expression of the vocal tract and larynx structure and function. An SNP-based voice profile of Nietzsche was inferred. The profile data was converted into bio-measures that were used to 3D-print a vocal tract and larynx through which phonation was organically generated. A composite of seven Text-to-Speech simulations was made using a sound morphing software. The result is presented in audio format and illustrates the first attempt at simulating the voice of a deceased person.

And some of the figures and tables are even better:

But there are some issues — and not just with the idea that there are such things as "voice SNPs" or that the "24 most informative" ones would allow a meaningful reconstruction of someone's voice, or that 3D-printing of a vocal tract and larynx would permit "synthesis" sounding like this:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

Nachdem Du mich entdeckt hast, war es kein Kunststück mich zu finden: die Schwierigkeit ist jetzt die, mich zu verlieren.
After you discovered me, it was no great feat to find me. The problem now is how to lose me.

The journal Investigative Genetics does exist, but doesn't have this paper available on its web site — although eager authors sometimes distribute preprints before the journal does. But the affiliation of the first two authors,  Flavia Montaggio and Patricia Montaggio, is given as the "Department of Forensic Genetics, University of Basel, Switzerland". And while the University of Basel certainly exists, it doesn't seem to have a "Department of Forensic Genetics", nor any personnel named Montaggio.

The affiliation given for the third author, Imp Kerr, is the Nietzsche-Archiv in Weimar; again, this institution does exist, but Imp Kerr doesn't seem to be employed there. Instead, according to Wikipedia,

Imp Kerr (born June 6, 1980, Uppsala, Sweden) is a Swedish-French artist living in New York City, mostly known for her fake American Apparel advertisement campaign. […]

Kerr spent her childhood and teenage years in Paris, France. She moved to New York in 1999, where she graduated from NYU with a degree in magazine journalism and a minor in philosophy. […]

In December, 2008, Imp Kerr created a set of architectural drawings portraying investment banks as Las Vegas casinos.

I have no idea whether anything in the Wikipedia article about Imp Kerr is true, but it's clear from internal evidence that the alleged Investigative Genetics article is a piece of performance art.  The author is unable to restrain a certain telltale playfulness, e.g. here:

Nietzsche’s vocalization exemplifies a flat, smooth, typical mild sounding voice: hued pitch, honeyed tone, low versatility in timbre, silvery hyponasal flow, disembodied texture with a touch of steel in vowels, and low head-nasal resonance (lower than expected in regards of Nietzsche’s robust mandibles). 

And there are many segments that display quasi-scientific terminology in meaningless or contradictory ways, e.g.

The diction is unified and follows a three-wave prosodic structure with light elasticity. Range goes from low C to the B above middle C. Fundamental frequency ranges between 156.8 and 157.5 Hz.

Nietzsche’s voice simulation was tested against a control database of 4,287 voice recordings of living German individuals. Comparative tests based on 526 VoiceRator parameters were performed and uncloaked 28 minor bias discrepancies between the control voice recordings and Nietzsche’s voice simulation, resulting in a 94.68 percent level of accuracy consistent with previous results.

Overall, I'm looking forward to the BBC coverage.



  1. Laura Morland said,

    March 20, 2015 @ 6:23 am

    Wow, after my initial disappointment, what a delight! I especially love the idea of a "Department of Forensic Genetics."

    Here Imp is quoted as saying: "I just like to create scenes and fictions, small worlds, situations that don't exist. And if possible, I make these situations complex, like if you look into them you should find more than you expect, but at the same time I care about making them minimal and appealing in appearance."


  2. Bob Ladd said,

    March 20, 2015 @ 6:37 am

    NB: Montaggio means 'construction' or 'assembly' in Italian, and in referring to cinema it's specifically used for the process of film editing. Seems unlikely that the authors selected this pseudonym by chance.

    [(myl) And the article was "Communicated by Margretta Covert, Yale University School of Medicine". Needless to say, there is no faculty or staff person at Yale Med named "Covert". But again, it's a fitting choice. Presumably there's only one real author, namely Imp Kerr.]

  3. Bill R said,

    March 20, 2015 @ 7:37 am

    I'd be impressed if they had reproduced a living person's voice, but this seems to be obviously fake.

    [(myl) It's not fake, it's art.]

  4. Colin Fine said,

    March 20, 2015 @ 8:29 am

    I'm reminded of the work of Joan Fontcuberta

  5. Rube said,

    March 20, 2015 @ 9:58 am

    While disappointed to learn this was a hoax (sorry, art), I 'm glad they picked Nietzsche instead of Jesus, otherwise, this story would be clogging my Facebook feed today.

  6. Nathan said,

    March 20, 2015 @ 10:08 am

    The phrases describing the voice are strongly reminiscent of that silly wine-tasting terminology.

  7. Rebecca said,

    March 20, 2015 @ 6:19 pm

    The references are fun to dig around in, too. She goes to some effort to cite existing journal volumes, but bogus papers, like her own paper on Nietzche, "Contesting Truth". The first thing that struck me as odd (other than the topic) was that the footer refers to this paper as Kerr et al, though she is listed as third author.

  8. D.O. said,

    March 20, 2015 @ 6:40 pm

    But departments of Forensic Genetics certainly exist. Only not in the University of Basel. E.g.
    Department of Forensic Genetics, Institute of Legal Medicine and Forensic Sciences, Charité-Universitätsmedizin, Berlin, Germany

    Department of Forensic Genetics and Forensic Toxicology, National Board of Forensic Medicine, Linköping, Sweden

    Granted, the majority of such establishments wear the badge of Laboratory rather then Department.

    The paper itself, though, is a complete disappointment. It does not reject the null hypothesis at the 5% level.

  9. ERV said,

    March 20, 2015 @ 10:37 pm

    I love how they managed to sneak 3D printing in here. All this is missing is epigenetic analysis of Denisovan or Heidelbergensis long noncoding RNAs followed by CRISPR mutagenesis.

  10. Randall said,

    March 20, 2015 @ 11:36 pm

    Imp Kerr has a genuinely fascinating blog at http://www.newshelton.com/wet/dry/ which I've read for a while now. this seems to be up her alley.

  11. El said,

    March 20, 2015 @ 11:42 pm

    The authors (or only Kerr) say the were funded by a real funding source (and a big and well known one), Swiss National Science Foundation. If this is a hoax, then Kerr should have chosen something different, as these big organizations do not usually play with this kind of stuff..

  12. Jason Eisner said,

    March 21, 2015 @ 1:52 am

    Impressed with the scientific imaginativeness and attention to detail — the artist knows the relevant science (and has a terrific ear for the conventions of scientific communication).

    This makes me wonder whether science fiction is managing to do the same. Sci-fi used to be mostly interested in space travel, life on other planets, physics, humanoid robots, cyborgs, occasionally AI or eugenics. But I haven't checked back since the early 1980's. Does contemporary sci-fi reflect contemporary science?

  13. Jason Eisner said,

    March 21, 2015 @ 1:57 am

    The Nietzche quote was carefully selected. That's Imp Kerr talking to us in Nietzche's voice.

  14. David Eddyshaw said,

    March 21, 2015 @ 3:55 pm

    I discovered the other day that Emil Kautzsch, the celebrated reviser of Gesenius' Hebrew Grammar, was a friend of Nietzsche's.

    I presume they bonded over spelling.

  15. DNA said,

    March 22, 2015 @ 3:50 am

    This is obviously nonsense. For example:

    > "Five different genomic DNAs were generated"

    There's no such thing as a genomic DNA. It's like saying that a money was spent. There's genomic DNA, there are genomic DNA libraries, but there are not genomic DNAs.

    More obvious nonsense:

    > Nietzsche’s genotype was singled out

    Hm, curious terminology.

    > after comparison to genotype data from one living relative of the Nietzsche family.

    Not a member of the family, but a relative of the family. Is that possible?

    > the phenotypic expression of the vocal tract …

    Say what?

  16. Listen to a “DNA-Based Prediction of Nietzsche’s Voice” | Movie News-Android-open education-Technology-news-news Sport said,

    March 23, 2015 @ 10:30 am

    […] her fake American Apparel advertisement campaign,” or so reads the Wikipedia page quoted by a Language Log post on the project. “I have no idea whether anything in the Wikipedia article about Imp Kerr is true,” […]

  17. Hear a “DNA-Based Prediction of Nietzsche’s Voice:” First Attempt at Simulating Voice of a Dead Person | Golden Gate Daily said,

    March 24, 2015 @ 2:13 am

    […] her fake American Apparel advertisement campaign,” or so reads the Wikipedia page quoted by a Language Log post on the project. “I have no idea whether anything in the Wikipedia article about Imp Kerr is true,” […]

RSS feed for comments on this post