In the Second World War, the Germans constructed a voice encryption machine. I have only very little information about this device. Can my readers help me to find out more?

In five weeks, at the Charlotte International Cryptologic Symposium, I will give a presentation about German ciphers in World War II. Of course, I will mention the Enigma, but my focus will lie on the less known German encryption systems. There are pretty many. I am aware of about a dozen machines, a dozen codebooks, over 20 manual ciphers, five encryption devices, and several steganographic methods the Germans used during WW2. I’m sure my list is not complete.


German encryption machines

The German encryption machines I will cover in my talk are mainly the ones mentioned in my blog post from 2016.


In this post, I mentioned that a journalist once had told me that the German WW2 general Erwin Rommel used a voice encryption machine in the tanks of the Africa corps. This sounded very interesting, but in the literature I was aware of this device was not mentioned. In a comment, blog reader Thomas Bosbach provided me a link to a TICOM document that mentions a speech encipherment apparatus. This device might be identical with the Rommel voice encryptor.


Voice encryption

Technically speaking, voice encryption is completely different from encrypting a text (most of the methods covered on this blog encrypt text). In WW2, voice encryption was still in its infancy. The Americans constructed a voice encryption machine named SIGSALY, which was as good as unbreakable, but extremely large and expensive. Only a few copies were built.


In the UK, Alan Turing built a voice encryption machine named DELILAH.


DELILAH was never used in practice. There was also a Russian voice encryption device. I have never seen a picture of it. The Germans constructed a device that automatically broke this Russian voice encryption. Here’s a photo of it (provided by Klaus Kopacz):


All in all, voice encryption didn’t play much of a role in WW2. The technology was still to immature to be really useful. For more information, I recommend the book How to Wreck a Nice Beach by Dave Tompkins (yes, the book is as good as its title).


The speech encipherment apparatus

The afore-mentioned TICOM document not only describes a speech encipherment apparatus, but also a so-called “Erschwerungsgerät” (aggravating device). This machine encrypted voice transmissions in a simple way. It was easy to break, but at least it prevented listening to a transmission with an ordinary radio device.

Here’s page 1 of the document:


The second page mentions that several companies and authorities in Germany worked on voice encryption devices during WW2. In 1943, the Wehrmacht tried to combine forces.


The third page contains a very brief description of the speech encipherment apparatus:


This document is the only one I know that covers German voice encryption technology in WW2. I have never seen a picture of a German WW2 speech encipherment apparatus, let alone a machine itself.

Does a reader know more about this topic? Any information  is welcome.

Further reading: A fascinating report of Enigma’s contemporary witness, Max Rüegger


Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Kommentare (13)

  1. #1 Rich SantaColoma
    13. Februar 2018

    Under a search for “erschwerungsgerat baustein”, This TICOM document comes up:

    … and it appears to be different than the pages you show in your post. So somewhere in there may be more information… I read through it and have not found it yet, perhaps someone else can. I’ll read the rest of it later. In any case, it is an interesting historical record.

  2. #2 Thomas
    13. Februar 2018

    Acoording to this TICOM report on German ciphony research ( pp. 36 – 43, see the conclusions on p. 43) there had been a lot of research during WWII, mainly by Oscar Vierling, but in the end Germany had no usable ciphony machine. The only apparatus ready to produce seems to have been the speech scrambler “Kleiner Baustein” (Little Building Block). According to the report diagrams of this device are (or were) available (footnote 102: “E-9”, maybe another report).

  3. #3 Thomas
    13. Februar 2018

    SIGSALY was the second US speech scrambling device after the A-3 system which was broken by German analysts, (this site contains more details about ciphony research).

  4. #4 Martin Gillow
    United Kingdom
    14. Februar 2018

    Cryptomuseum have info on voice scramblers inc. SIGSALY but not too much that early. If you want info on Lorenz / Colossus for your presentation, I have written a website with an online fully working simuator of Lorenz SZ40/42, Colossus and Dragon cribbing machine if that’s any help. &

  5. #5 Jerry McCarthy
    England, Europa.
    14. Februar 2018


  6. #6 Rich SantaColoma
    14. Februar 2018

    I think this document gives the most detail, and possibly the final word (because of the great and detailed descriptions of the machines) on all the systems discussed:

    Starting on page 38, it discusses all the systems found and known, relating to voice encipherment by the Germans during WWII… known as “ciphony”. It describes the methods of inversion, “wobbling”, “building blocks” large and small, and more.

    After describing, in detail, all the known methods (including the ones mentioned in the blog article), it concludes (page 45): “Germany had no usuable ciphony machines; and Germany probably would not have had any usable ciphony machines even if the war had gone on several more years.”

    But I also find it interesting that the blog paper explains that the “… control of the “Wobbelung” is carried out by the ten peg wheels of the SZ42C.”

    Since the SZ42 was a text machine, I wondered if the machine was adapted to encipher analog audio information… the “blocks”… through the same wheel circuits? I mean, they may not have built a unique device to scramble the blocks, but used the switching capabilities of existing ones?

    In any case, when reading the descriptions of the methods in the paper I link here, it is no surprise… given the complexity of the problem, and the attempted solutions to them, combined with the limited audio resolution capabilities of the technologies in the 1940’s: microphones, wire recorders, so on… it is no surprise that these methods failed. I would say they simply didn’t yet have the necessary tools to make it work.

  7. #7 Thomas
    14. Februar 2018

    There is a very interesting TICOM report about “Laboratorium Feuerstein”. Burg (castle) Feuerstein was Oskar Vierling’s laboratory where he developed several SIGINT devices on behalf of the Wehrmacht. Esp. here (pp. 41 and 42) is provided detailed information about speech scramblers: According to the references given on p. 82 there are “Detailed Feuerstein Technical Project Reports” E-9, E-10 and E-11 concerning ciphony. Unfortunately the TICOM archive contains only E-8 but not the rest of the Feuerstein reports E-7 – E-17. Up to now I haven’t managed to find these reports online. Maybe somebody else can find them.

  8. #8 Siegfried Zimmer
    14. Februar 2018

    I assume you mean Erwin Rommel.
    Manfred Rommel was his son and Mayor of Stuttgart from 1974 until 1996.

  9. #9 Klaus Schmeh
    14. Februar 2018

    @Siegfried Zimmer: You’re right, I corrected it.

  10. #10 Thomas
    14. Februar 2018

    About the voice encryption system used by Rommel’s forces in North Africa (developed not by Vierling, but by Telefunken):

    “Speech privacy system:

    From TICOM I-46 ‘Preliminary Report on Interrogation of Dr. Otto Buggisch (of OKH/Gen.d.NA) and Dr. Werner Liebknecht (employed by OKH and OKW as tester of cryptographic equipment) 23  June 1945’ , p3

    Voice cipher machines: 
    C. Superimposing noise of electrons on speech and taking it out by 180 degree phase; A fixed noise level was not at all successful, but a variable number when employed had some small degree of success. This plan was actually employed back in the North African Campaign when communication from Athens, Greece to Derna, North Africa via Crete was maintained. A noise level of 1 to 4 was employed; however, the problems involved outweigh the advantages derived and the equipment was destroyed by fire and never replaced.

    From  ‘Spread spectrum communications handbook, Volume 1’  , p49 [Reference found via historykb user eunometic]

    The Germans used Kotowski’s concept as the starting point for developing a more sophisticated capability that was urgently needed in the early years of World War II. Gottfried Vogt, a Telefunken engineer under Kotowski, remembers testing a system for analog speech encryption in 1939. This employed a pair of irregularly slotted or sawtoothed disks turning at different speeds, for generating a noise-like signal at the transmitter,to be modulated/multiplied by the voice signal.The receiver’s matching disks were synchronized by means of two transmitted tones, one above and one below the encrypted voice band. This system was used on a wire link from Germany, through Yugoslavia and Greece, to a very  and/or ultra-high frequency (VHF/UHF) link across the Mediterranean to General Erwin Rommel’s forces in Derna, Libya.”


  11. #11 Christos
    15. Februar 2018

    Gottfried Vogt’s WWII voice encryption system should be the one described in the patent:

  12. #12 Thomas
    15. Februar 2018

    Vogt’s patent linked under #11 is a sawtooth signal generator which afaics has nothing to do with voice encryption.

    Vogt’s system used by Rommel’s forces was based on superposition of noise created by rotating irregular slotted or sawtoothed disks. The underlying principle was obviously the same as in the patent his boss at Telefunken, Paul Kotowski, had previously applied for:

  13. #13 Klaus Schmeh
    17. Februar 2018

    Thank you very much to all! This is a lot of material. I will look through it and use it for my presentation. Maybe, I will blog about it.