Marsh warbler mimicking identified African and European bird sounds in recording
A recording of a Marsh warbler has been disassembled into separate sequences representing sounds made by both African and European bird species. The recording has been split into over 500 separate sound clips, being downloadable here as a single Audacity file. 14 African bird species have so far been identified in the recording and each identified sequence is presented in the file.
The first species identified to mimic bird sounds from its wintering quarters
The Marsh warbler, Acrocephalus palustris, is one of the world’s best mimics. Many other birds mimic sounds in its surroundings, but the Marsh warbler was the first that was found to mimic bird sounds both from its summer and its winter quarters. In my opinion, its performance outshines all other birds when it comes to song. It's been studied by Françoise Dowsett-Lemaire who herself identified or found information in the literature on 100 European bird species [1-3] and then herself identified 109 African bird species  being mimicked by the Marsh warbler. Her paper is, to my knowledge, the only comprehensive study made on this subject, so there is definitively room here for further investigations by amateurs and professionals alike that really contribute to our knowledge. The paper  includes a list of African bird species identified and also contains some information about what type of sounds are being mimicked, so this can be used as a guide for anyone interested in identifying African birds being mimicked. In order to identify the African birds being mimicked, it's of course necessary to be very familiar with the sounds of European birds, in order to initially eliminate these. Unfortunately, the papers were published as regular paper documents only, and the recordings that she based her studies on are not available. Françoise Dowsett-Lemaire has also co-authored a paper  on a hybrid between a Marsh warbler and a Blyth's reed warbler, and a recording of this bird is available, with information on respective bird species being mimicked identified, and this gives further leads.
Do other excellent European mimics also imitate bird sounds from winter quarters?
Other European birds that excel at mimicking are the Bluethroat  (Luscinia svecia), the Icterine warbler  (Hippolais icterina) and the Marsh warbler's close relative Blyth's reed warbler  (Acrocephalus dumetorum). Scientists have studied the song of these species too, but no comprehensive study of whether they mimic birds from their wintering quarters exists. The Blyth's reed warbler is said  to mimic at least one bird species, the Red-wattled Lapwing (Vanellus indicus), from its wintering quarters though, so anyone having a thorough knowledge of both European and south east Asian bird sounds may be able to make interesting identifications. Interestingly, the Bluethroat, the Icterine warbler, the Marsh warbler and Blyth's reed warbler are all known to also mimic each other. Recordings also exist of the Bluethroat mimicking mobile phones and cow bells, and the Icterine warbler has been interpreted as mimicking gambling machines .
Marsh warbler and Blyth's reed warbler engaged in interspecies antiphonal song
I did have a wonderful experience one June morning at Järfälla kyrka in the outskirts of Stockholm, when I realized that a Marsh warbler and Blyth's reed warbler sitting within 10m of each other were engaged in antiphonal song, with the Marsh warbler leading and the Blyth's reed warbler following. I didn't consider it at the time, but as the Marsh warbler and Blyth's reed warbler winters in different parts of the world, their shared vocabulary can only be constituted by mimics of European bird sounds, and as far as I can remember they did indeed mimic European birds only. How the Marsh warbler knew that Blyth's reed warbler couldn't follow if it mimicked African birds, I don't know, but it still seemed to have figured that out. They did however stick to their respective characteristic singing styles, so there was no doubt about who was who. The Marsh warbler rapidly mimicked a sequence of a few species, and then politely waited for a response from the Blyth's reed warbler. The Blyth's reed warbler replied by mimicking exactly the same sequence of species, but much more slowly and repeating each sound several times, as it typically does. Unfortunately, I didn't have a recorder at the time.
Using Françoise Dowsett-Lemaire's publication as a guide, I've made an attempt to identify African birds being mimicked in a recording I made in 2014. According to Françoise Dowsett-Lemaire, recordings should be at least 45 min long in order to contain the full vocabulary of an individual warbler, but I've still been able to identify 30 European species and 14 African species in the just 19 min long recording. The full continuous recording is freely available as an .mpg file from Xeno-Canto at www.xeno-canto.org/195477, as long as you credit the author if redistributing the file. The recording was made on Kökar in the Åland archipelago spanning the Baltic sea between Sweden and Finland. Road traffic on the tiny island is very limited early in the morning, so background noise is hardly noticeable at all. The recording was made using a Sony Xperia mobile phone, so the intrinsic noise is a bit high. I simply sneaked as close to the bird as I dared, and left the phone recording at a distance of around 5m from the bird. I've since bought a Sony PCM-M10 so I hope to be able to make better recordings next spring. I've made a simple noise measurement of the internal microphones of the Sony PCM-M10, and they seem to have a SNR some 17-19dB better than the Xperia.
Method used to disassemble recording
The method I've found to be most effective is to use mixer software and split the single recording into tracks, each track corresponding to a single species being mimicked. Each track then contains a sequence of clips, each clip corresponding to a single continuous sequence of mimics. When all the clips split between tracks are played, the result sounds identical to the original recording. One can then choose to play only selected tracks, for example just tracks containing yet unidentified sounds. The recording is at present split into 44 tracks corresponding to the number of species being mimicked that have been identified, two tracks containing silence and disturbing click noises respectively, so you can select not to listen to these, 19 tracks where unidentified sounds that recur are found, and finally one track where remaining yet unidentified non recurring sounds still remains. The tracks are named in accordance with the species being mimicked, the African birds using Latin names and the others so far using Swedish. I initially used Mixpad, which has a nice multi coloured interface, but Mixpad is not freeware and one does after an initial test period have to buy the software. I then changed to the freeware Audacity, and it seems to be at least as good as Mixpad, but free of charge. Also, for users familiar with software programming, you can make your own functions and add to the original functionality, so I would recommend anyone who wants to do a similar effort to use Audacity, and it can be downloaded by clicking this link. Also, the Mixpad file format seem to be readable by that software only, and transferring these files to any other mixer file type is very cumbersome indeed, possibly intentionally. A 105MB .zip file can be downloaded here, that after unzipping may be opened in Audacity. It is presently divided into over 500 clips, and although most of the European bird sound identifications are reasonably reliable, the African bird sounds identified are much less so. However, seven of the clips containing mimics of three African bird species have kindly been confirmed as being correct by Françoise Dowsett-Lemaire, the foremost expert in the field.
For the actual identification process, a comparison has to be made between recorded sounds of relevant bird species and the sound made by the Marsh warbler. Audacity may present the recording as spectrograms, which speeds up visual recognition when looking through the recording, but for actual identifications audile comparison is superior. Experienced European ornithologists would perform this by heart and only rarely refer to reference recordings, but when it comes to identifying African bird sounds reference recordings are necessary. The best such recordings are commercially available as CD:s or mp3-files, and two that have been recommended are R. Stjernstedt's recordings of birds of Zambia and C. Chappuis recordings of birds of West-Central Africa. When it comes to freely available reference recordings, Xeno-Canto's collection is the most comprehensive I've found, but the constributors are primarily not experts and misidentified recordings exist in their collections. Another good source is the Macaulay Library at the Cornell Lab of Ornithology, it seems to have only professional contributors, but is not as comprehensive as Xeno-Canto.
Further studies and questions
Please do feel free to download the file, and if you have any input,
such as corrections or new identifications of species you are most welcome
to contact me via the contact form available at Xeno-Canto,
easily found by clicking at the link to the original file stored there.
Also, if you have a recording of your own where you wonder about some
sound that may or may not be a mimic of an African bird, you are of course
most welcome to contact me, perhaps I might be able to help you out. Please
notice that the file constitutes work in progress, and there are most
likely lots of misidentifications, so don't expect a complete and fully
correct disassembly of the recording. If you have any other information
on European birds mimicking birds from their non-European wintering quarters,
that is also of interest.
 Françoise Dowsett-Lemaire, “Vocal behaviour of the marsh warbler,” Le Gerfaut 69, (1979), pp. 475-502. ISSN 0016-9757. In English, with summaries in French and Dutch. Not freely avaliable, so you need to pay for downloading it or order a paper copy to your local library.
 Lemaire, Françoise. 1974. Le chant de la rousserolle verderolle (Acrocephalus palustris): étendue du répertoire imitatif, construction rythmique et musicalité. Le Gerfaut 64: 3-28. In French, with summaries in English and Dutch. Not freely avaliable, so you need to pay for downloading it or order a paper copy to your local library.
 Lemaire, Françoise. 1975. Le chant de la rousserolle verderolle (Acrocephalus palustris): fidélité des imitations et relations avec les espèces imitées et avec les congénères. Le Gerfaut 65: 3-28. In French, with summaries in English and Dutch. Not freely avaliable, so you need to pay for downloading it or order a paper copy to your local library.
 The imitative range of the song of the Marsh Warbler Acrocephalus palustris, with special reference to imitations of African birds. Ibis 121: 453-468. Dowsett-Lemaire, F. ISBN/ISSN: 00191019. In English. Not freely avaliable, so you need to pay for downloading it or order a paper copy to your local library.
 Hybrid Marsh x Blyth's Reed Warbler with mixed song in Finland in June 2003 by Antero Lindholm, Staffan Bensch, Françoise Dowsett-Lemaire, Annika Forsten & Hannu Kärkkäinen (Dutch Birding 29: 223-231, 2007). In English. Freely readable document and freely downloadable sound recording from http://issuu.com//dutchbirding/docs/db_29_4_2007
 Wallschläger, D. (1978). Imitationsleistungen eines Blaukelchen, Luscinia svecica (L.). Mitt. Zooll. Mus. Berlin 54 (Suppl.), Ann. Orn. 2 : 173-181. ISSN 0232-5519. In German. Not freely avaliable, so you need to pay for downloading it or order a paper copy to your local library.
 Vocal mimicry in the song of the icterine warbler, Hippolais icterina (Sylviidae, Passeriformes) Zuzana JUZLOVÁ and Jan Riegert, Folia Zool. – 61 (1): 17–24 (2012). In English. Freely avaliable for downloading from www.ivb.cz/folia/61/1/FZ_011.pdf
 Biology Bulletin, December 2010, Volume 37, Issue 8, pp 846-860, 24 Dec 2010, “Individual, population, and geographic differentiation in advertising song of the Blyth’s reed warbler, Acrocephalus dumetorum (Sylvidae)”, I. M. Marova, V. V. Ivanitskii, O. D. Veprintseva. In Russian. Biology bulletin of the Russian academy of sciences : a translation of significant articles from Izvestija Akademii nauk - Ser. biologiceskaja. ISSN 1062-3590 has en English translation.