An informant and pedagogue, who knows both Chinese and Russian, claims:

Pinyin is based on a Russian transliteration system and on the conventional method of transliterating Russian. Hence, the existence of such graphs as "x" which is actually the Cyrillic letter used for the palatal spirant involved. "Zh," in turn, is the usual transliteration for the Cyrillic letter employed for the unaspirated alveolar affricate. "C," in Slavic languages that use the Latin alphabet, represents "ts." The use of the voiced consonant graphs for unaspirated consonants also corresponds to the Russian usage. It is interesting to note that the earliest versions of pinyin actually used one or two Cyrillic letters. Wade-Giles, the oldest of the systems, has undergone several modifications in its long history. Certain of its seeming peculiarities reflect the fact that it was originally intended to provide a bridge among dialect versions of Mandarin. It is, however, linguistically more accurate than either pinyin or Yale in using only the graphs for unvoiced consonants (Mandarin has only five voiced consonants, m, l, n, r, ng) and indicating aspiration by means of an apostrophe.

Q: I'm convinced. What do you think?
A: If a complete innocent read "Zhu Rongji" in the newspaper and then heard television or radio newsreaders pronounce it as Жｕ Ｒｏｎｇжｉ, it would be a logical guess that the Pinyin zh is Russian in one way or another. The informant, however, knowing as he does both languages under discussion, is no such innocent. The zh in Brezhnev has nothing to do with the Pinyin zh, and Pinyin x sounds nothing like Cyrillic Russian х. What he writes is rife with ambiguity, non-sequiturs, and as a whole is extremely mal-informative. But on the surface it is all very convincing.
Q: I want reasoned evidence, not assertion.
A: Ask on then. But in order to parse his claims, you'd best start with some basic questions.
Q: What's the difference between transliteration and transcription?
A: Loosely, transliteration covers both terms. Less loosely, transliteration means a letter to letter correspondence (indifferent to pronunciation per se) between scripts (i.e. writing systems) that use an alphabet. Transcriptions on the other hand attempt to record/reflect pronunciation from languages without a script, or whose script -- as in the case of hanographic Chinese -- is not alphabetic. Transcription can either use a more or less conventionally limited set of symbols (Latin, Cyrillic, Arabic etc. letters) or that more expansive and discriminating tool, the IPA. The Russian genitive ending -ого is transliterated into English as ogo, but since the г is here pronounced v, a transcription of it into English would be ovo. Or supposing that Zuyin Fuhao were used in transcription or transliteration of Latin scripts and there occurred the name "Mann", a transcription would be ㄇㄢ and a transliteration would be ㄇㄚㄣㄣ. To exaggerate, transcription is concerned with sounds, and transliteration is concerned with letters. The various romanisations (in so far as they aren't considered scripts or proto-scripts) are of course transcriptions of Chinese. We wish the informant had been clearer about what he meant by "transliteration", it would have made it easier to understand his claim. Since he didn't, we'll have to examine a wide spectrum of impossibilities.
Q: ... What's another basic question?
A: Ask what 尖团 is. This will be central to the matter of "dialect bridging" and what Wade-Giles is or isn't.
Q: What is 尖团?
A: 尖音 and 团音. In a purely hypothetical Pinyin, it would be the difference between, for example, the spellings 'zian' and 'gian'. Some people translate jianyin and tuanyin into Englishas "sharps" and "rounds".
Q: I've never seen such spellings as as 'zian' and 'gian'.
A: Right you are. ... The people who know best decided that such a distinction should be (apart from the sung portions of Peking Opera of course) brutish and uncultivated, while distinguishing retroflexed shi and non-retroflexed si on the other hand should be the very hallmark of good breeding and talking proper.  ... Anyway, s/z/c followed by i or umlaut u would be jianyin and g/k/h followed by i or umlaut u would be tuanyin.
Q: Such sounds exist? Other than in Cantonese or something like that?
A: They occur in some areas of the North.
Q: What's the relation between those spellings and actual pronunciation?
A: Algebraic. Depending on which area in the North where such a distinction exists, one of the two sets might be pronounced like Pinyin j/q/x, and the other pronounced as it looks to be. The point is, regardless of how the sounds are actually realised, 酒(z-) and 九(g-) are not homophones. Pekinese j q x is a confluence of two originally separate sets of sounds. EFEO, the romanisation presently near death in French, handles Pinyin's j q x in 尖团 manner, as ts ts' s on the one hand and k k' h on the other.
Q: What happened around 1955?
A: A coup d'etat in language policy. "Putonghua" was re-defined to mean exactly the same thing as the Guomindang's "Guoyu", i.e. something based strictly on the phonology of Peking, with 尖团 differentation and some vowel distinctions (-e / -o) declared sub-standard. At the same time, since this new Pekinised Putonghua was declared the standard language of the entire Han Nation, non-Northern languages spoken in such areas as Shanghai, Canton, Fujian etc., became correspondingly demoted. Really it was simply an adoption of GMD language policy.
Q: What was Beila?
A: One of Pinyin's two "parents". Beila is the acronym for Beifangxua Latinxua Sin Wenz 北方话拉丁化新文字, (incorrectly also known in English simply as Latinxua; incorrect because there were Latinxuas for areas other than the North). Beila was based on a broadly conceived Northern. The other, less apparent but more significant parent of Pinyin was Gwoyeu Romatzyh, which was based squarely on Pekinese phonology ... Now you can ask non-basic questions.
Q: Can Pinyin's "use of the voiced consonant graphs [i.e. b d g j z] for unaspirated consonants [i.e. the sounds ㄅㄉㄍㄐㄗ]" be attributed to Russian influence?
A: If so then this same usage of b d g (plus, here and there, j and z) in romanisations from 刘孟扬's「中国音标字书」in 1908 to Lessing-Othmer's 1912 system for German speakers, to Gwoyeu Romatzyh, Yale, and the GMD rump-state's 1986 「国语注音符号第二式」, have all been prey to the influence of "Russian transliteration". That would be quite a scandal.
Q: Why use b d g in such a manner?
A: Because in the language whose romanisation is being discussed, phonemic contrast is more like English "pack" vs. "back" than French "pas" vs "bas". De-voice the b in "back" and no one will even notice; de-aspirate the p in "pack" and you'll be understood to have said "back". And in French just the opposite -- de-voice the b in "bas" and you'll be understood to have said "pas". What is phonemic (i.e. semantically differentiating) with Northern/Pekinese stops and affricates (Pinyin p t k and q c  vs.  b d g and j z) is aspiration vs. non-aspiration, not voicing and non-voicing. Further, to talk of "voiced consonant graphs" as the informant does, is completely unacceptable. Graphs no more jump up and speak than rocks do. Sound values of letters are assigned by humans, humans speaking languages with different sound systems and thus different needs. Following the informant's logic, the p in English "pack" is also wrong, a true p is of the Spanish/French type, and "pack" should be spelled "p'ack". But that would be un-economical, as are Wadian apostrophes.
Q: But what about accuracy?
A: Phonetic/phonemic scripts (i.e. writing systems) are a bundle of conventions using a fairly limited set of symbols; likewise with transliterations and (non-IPA) transcriptions. From these conventions you can ask economy of letter use, rule regularity, internal non-ambiguity, predictability of pronunciation, aesthetics, or even similarity to letter-use of a certain language ... but not a universal "accuracy". Catalan x is an English sh; German z is a ts; Albanian q is a semi-palatalised ch; Turkish c is a j. Who is to say that these are inaccurate? The question of accuracy per se arises only in using the IPA. This didn't exist in Mr Wade's time. If it did he might have used it to spare him all the contortions of apostrophes, umlauts, and various hats on vowels. Furthermore, it's strange that anyone would speak of Wadian accuracy in any practical sense given that outside of refined use, its apostrophes and umlauts are usually missing, lending it a such a high degree of ambiguity that there results an "inaccuracy" not at all requiring inverted commas. Is a Wadian "chun" what it appears to be,thus in Pinyin, zhun? Or have apostrophes and hats been lost, such that (in Pinyin) it is really chun, jun, or qun? There's nothing wrong in "liking" Wade-Giles over other systems, but to rationalise one's subjective preferences by talking about accuracy is another matter entirely. Anyway, the informant must be limiting his remark to pedagogy, because the question of phonetic accuracy doesn't arise for the great mass of passive consumers of romanisations: those that only need to read and write them. For them what is desirable is that graphological forms are not constantly changing over time (e.g. from Morrison to Wade-Giles to Pinyin) and in space (e.g. from Wade-Giles to EFEO or Lessing-Othmer). Sound is not so important; what is problematic is that Amoy and Xiamen are the same city and that I King, I Ching, Yi Ching, I Ging, I Djing and Yijing are all exactly the same book.
Q: Is there no situation in which Pinyinesque unvoiced b d g would be inappropriate?
A: Certainly. The point is that it depends on the language. (In Malay it would be impossible, not just "inappropriate".) In transliterating a language like Sanskrit, where aside from ㄅ and ㄆ there are voiced versions of both, so that p ph b bh is the convention, with the h serving the same aspirating function as the Wadian apostrophe. But even so, this convention of using h to indicate aspiration is not sacrosanct accross languages. In Y R Chao's 「通字罗马字」-- an interdialectal romanisation where three rather than four of these phonemic contrasts occur -- the problem of voiced b and d comes up; he adds an h for voicing: bh and dh. Very different to the principle applied in Roman transliteration of Indic languages. But all this still has to with convenience, economy (simpler graphs for more frequent sounds, more complex graphs for the more infrequent ones) and so forth, not "accuracy". As long as it is made clear what letters represent what phones or phonemes, that is sufficient accuracy. For example, to language-learners, it's necessary to explain and demonstrate that Wadian "sh" is a retroflex; simply looking at it, there's no reason to assume it's anything other than an English sh. Oughtn't the informant consider this also a lack of "accuracy"?
Q: Is it true that "Certain of [Wade-Giles'] seeming peculiarities reflect the fact that it was originally intended to provide a bridge among dialect versions of Mandarin"?
A: We wonder what he means by Wade-Giles, and about his conception of the temporal continuum of romanisation in general. Is it limited to the prosaic needs of American pedagogy i.e. Pinyin, Yale, Gwoyeu Romatzyh, and a vague "Wade-Giles"? Between 1605 and 1958 there were over 45 romanisations done by persons from either side of Euroasia. Functionally -- despite combination-use and/or historical atrophy -- they can be schematically divided as follows:

1) Those intended as a learning device for non-natives;
2) Those used only for practical and limited ends such as romanisation of place names;
3) Those intended for or by sinophones as replacement of or as auxiliary script to hanzi.

The Church Romanizations, Gwoyeu Romatyzh, and Beila fall into the third category. We wonder if the informant is using Wade-Giles as a cover term for any and every thing devised by Englishmen in categories (1) and (2).
Q: "Church Romanisations"?
A: The Protestant missionaries were fervent in their desire that converts should be able to read the Bible in their own vernacular. (Of these systems, the only one left, with a dwindling constituency in Taiwan, is the one for Xiamenhua. In the 1890s persons from southern Fujian and Taiwan would travel throughout the hinterland arousing comment by the fact that although they were unable to read or write hanographs, they were fully literate in this system.)
Q: I asked about Wade-Giles and intended dialect-bridging.
A: Morrison, Wade, and Postal form a continuum. Morrison differentiates aspiration/non-aspiration and 尖团 in the same manner as EFEO, and has very English spellings like ee and ow for later Wadian i and ou. Wade modified Morrison by making such spellings less parochial; and, much more importantly, by abolishing 尖团 in favour of Pekinisation. The Postal System is more of an Collection than a System. The apostrophe was not welcome with the English reformed drug dealers in charge of Posts and Telegraphs, so it was done away with. (Apparently no one thought to replace it with the h used in many of the Church Romanisations, or they did but spellings like "chhing" [= Wade-Giles ch'ing] were found unsightly.) The eclectic result retains various aspects of Morrison but has no apostrophes (i.e. it does not distinguish aspiration/non-aspiration), and for some place-names in the Southeast uses transcriptions based on Cantonese or Fujianian, sometimes mediated by old French ("Canton") or Portuguese ("Quemoy") spellings. At the last published modification of Wade-Giles in 1912, Standardised Putonghua/Guoyu was mercifully still a thing of the future, so if by dialect-bridging he means for example the fact that Pinyin e is usually Wade-Giles o but sometimes Wade-Giles e (shang-k'o vs. k'e-jen  上课/客人）, or has peh (Hopeh) and po (Li Po) for modern standard bei and bai, this no doubt results from prevalent pronunciation at the time even in Peking [the e ultimately resulting from atrophied 入声 influence i.e. glottal stop after the vowel?], not some intended dialect-bridging. On the other hand, if he means Sinkiang, Tsingtao, Tsinan, Chungking, Peking, Amoy, Quemoy ... which are indeed peculiar in relation to the actual Wade-Giles spellings Hsin-chiang, Ch'ing-tao, Chi-nan, Ch'ung-ch'ing, Peh-ching, Hsia-men, Chin-men ... then he's confusing Wade-Giles with the Postal System. ... you will notice that before i, Postal never has ch, only k ( = hypothetical pinyin g,k) or ts ( = hypothetical pinyin z,c). So in answer to the question about intended dialect-bridging, if by "Wade-Giles" the informant means Wade-Giles, there was clearly no such intention, in fact, just the contrary. If by "Wade-Giles" he means the Postal System, then yes there is a little of that function but it arose from practical need or convenience, not intent. If he means Morrison, the answer may be yes.
Q: Well then, is not a fact that "C, in Slavic languages that use the Latin alphabet, represents ts."?
A: It's incontestable. But so? This just shows that there is a ready-made convention in scripts using the Roman alphabet for this use of c. These languages have a lot of the ts sound, as does Chinese; it makes sense to so use the letter. But what is this doing in a paragraph starting out, "Pinyin is based on a Russian transliteration system and on the conventional method of transliterating Russian."? The scripts used by these languages (of which Magyar is not Slavic anyway) are neither transliterations of Cyrillic written Russian nor transcriptions of oral Russian, and thus have nothing to do with the thesis. Even German c (as in the very name of the letter) is sometimes a ts; More "Russian transliteration"?
Q: Then what about what he says about Pinyin x and zh?
A: But what "Russian transliteration system" (or transcription system?) is he talking about? Roman English transliteration of Cyrillic Russian? Cyrillic Russian transcription of Pekinese? If he means Roman English transliteration of Cyrillic Russian, then since ж→zh, ч→ch, ш→sh, ц→ts, х→kh, then not only Pinyin zh but also ch and sh are Russian derived, and we'd wonder why Pinyin uses c instead of ts, why it uses h instead of kh, and why it writes ㄖ as ri instead of zhi! If he means Cyrillic Russian transcription of Pekinese, then we'd point out that Cyrillic Russian ж (like with the Wade-Giles/EFEO/Lessing-Othmer French-type j) is the transcription not of ㄓ but of ㄖ, and that the Cyrillic Russian transcription of Pekinese ㄒ  is not х but с (= s)! Or perhaps he's confusing Russian transcription of Chinese with how cyrillicised Dungan is written; but even then he's still very much at sea. It's a small miracle he doesn't somehow ascribe Pinyin z to "Russian transliteration" as well.
Q: Then, how are Pinyin zh and x to be explained?
A: Zh, very simply and obviously by the inner logic of the system. Z c s are dentals; add an h after them and they become retroflexes. (And since many speakers don't differentiate the two sets of sounds, this is also a very reasonable treatment.) In other words, just as the ch is derived from the c and the sh is derived from the s, so is the zh derived from the z. As for Pinyin x, if Portuguese x can be an sh sound, Pinyin x can certainly be ㄒ.
Q: So the Pinyin-Cyrillic Russian connection is mythical?
A: Yes. But what is very strange indeed is that the informant doesn't trace the supposed Cyrillic Russian source of Pinyin -- especially the x -- back to Beila, as was the wont of that earlier generation of GMD court-linguists on Taiwan. This would have been a much stronger argument. Or perhaps his 'Pinyin' is as broad a category for him as 'Wade-Giles' may be. Anyway, since he doesn't make the argument, you should do it for him.
Q: Give some more details about Beila.
A: Beila was devised in 1929 by Qu Qiubai .... yes, with input from Russian linguists ... and revised/finalised at a conference in Vladivastok in 1931. Some of the changes made in 1931 were:

h (used as an initial, not retroflex-marker) → x
jh → rh
an', en', in' → ang, eng, ing
on → ung

None of these changes represent an increased 'Cyrillic' influence. Use of x instead of the original h, may seem to, but we shall see below that this is not the case.

Aside from the Cyrillic Question, if the informant wants to extol a dead or dying system, especially considering the "dialect-bridge" criterion, he should choose Beila, not Wade-Giles. Based on broadly conceived Northern, rather that just one locality (Peking), 1) Beila differentiated Pinyin e into o and e (much like Wade-Giles, EFEO and Lessing-Othmer) and differentiated Pinyin j q x along 尖团 lines, as z/c/s and g/k/x; 2) Where Pinyin has zi ci si zhi chi shi ri, Beila wrote no vowel; 3) Beila had neither mma mar maa mah tonal-spellings nor tonographic ā á ǎ à ē é ě è ī í ǐ ì ō ó ǒ ò ū ú ǔ ù;it just didn't indicate tone; 3) On the other hand, it did contain some disambiguating special spellings -- non-tonographic Pinyin's "Shaanxi" for 陕西 is so derived. In connection with point (1) above, it's important to bear in mind that rather than representing a pronunciation standard, Beila was a script, the exact pronunciation of which could differ from speaker to speaker, much like Nynorsk. So for example, the actual pronunciation of z c s g k x when followed by high vowels ( ㄧ and ㄩ ) would depend on the speaker; some would pronounce them all as Pekinese Pinyin j q x, others would differentiate the two sets in one way or another. Similarly for o/e. ... It's really no more strange or exotic than the fact that Australians pronounce the d in "duel" as a j, or that speakers of Irish, American, Australian, and RP English have four different pronuciations of the word "car"; the graphological form is unchanging ... So, Beila did not have Pinyin's j q x, but rather used z c s g k x before both high vowels and low vowels. In other words, the letter x was not limited to what the informant calls the "the palatal spirant involved". Pinyin "houlai" was Beila "xoulai".
Q: Clear, evident and undeniable borrowing of a Cyrillic letter! What you said about the Pinyin x was convincing but with Beila it's another matter altogether!
A: Is it?
Q: Otherwise why use x?? Why not use h??
A: A parochial but logical deduction you've made on behalf of the informant. 1) Again, intra-system logic: Why not use h? Because it was decided in 1931 to not use h as an independent letter but only as to indicate retroflexion of z c s (and to differentiate ㄖ from ㄦ). If h had been used both as a symbol to indicate retroflexion and as a letter in its own right then a spelling like "shu" would be ambiguous; it could be either double-syllable s + hu [似乎] or single-syllable sh + u [书]. Keeping the two functions separate, as finalised Beila did, there is no confusing double-syllable "sxu" and single-syllable "shu". 2) Neither Cyrillic Russian nor Roman English has a monopoly on the graph form x or the sound value thereto assigned. It has all sorts of sounds. Aside from Greek χ and Cyrillic х pronounced ㄏ, so is IPA [x] as well as the x in some words in or from Spanish, like "Mexico" and "Don Quixote". The letter x has an sh value in Catalan, Maltese, Basque, and often in Portuguese. In fact, the first three romanisations of Northern Chinese, of which Ricci's was one, used x for what is today written sh in Pinyin. In Albanian, the letter x is a dz sound. So it's not as though God ordained that x can represent nothing but ks. 3) Although not of phonemic importance, from the mouths of Northerners ㄏ is indeed an IPA [x] not [h]. (Another letter-use where Wade-Giles does not and need not stand up to the test of ill-conceived accuracy; wouldn't kh have been more "accurate"?) Further, we might ask the informant if Y R Chao's Beilaoid use of x in 「通字罗马字」(where 希 is xi, 花 is xua, and 忽 is xut) is also "based on a Russian transliteration system". Nor can, as the uninformed might think, Beila's y be attributed to Cyrillic Russian influence: IPA [y] is ㄩ (and so generally is the y found in written Norwegian, Dutch etc.); Russian у on the contrary is IPA [u]. ... If preposterous "Russian transliteration" etiologies are the order of the day, we would suggest that the Gwoyeu Romatzyh use of y (Pinyin zhi = GR jy) is a more fitting candidate: in Russian to English transliteration is it not the case that ы → y ? ...For Pinyin's y, Beila used (but only to disambiguate syllable-boundaries) j, which again is the IPA symbol. Analogous to the switch from h to x in 1931, jh was revised to rh so that the function of j would be delimited and unambiguous.
Q: Well then, Beila wrote Pinyin zi ci si zhi chi shi as z c s zh ch sh, with no vowel at all. Certainly this, at last, is a clear sign of Russian transliteration.
A: The informant should feel very indebted to you. Polish, which we emphasise is not a "transliteration of Russian", has words such as "w" and "k", and the Church Romanisation for Ningbohua wrote the syllable ㄙ simply as "s", thus “思想” as "s-siang". Take out that hyphen and it's the same as Beila. With some 14 romanisations proposed or in use (not including the Church Romanisations, which had different letter-use from area to area) between 1890 and 1914 it's really not very clever of anglocentrists to think that everything they don't like has something to do with the Bolshies. C (for ㄘ) was already used by the aforementioned 刘孟扬, and both z and c (for ㄗ and ㄘ) by another Mr Liu in 1914. Others used g, q or even v (from the Greek, ν = n) in place of the bothersome double-letter ng.
Q: But ... "it is interesting to note that the earliest versions of pinyin actually used one or two Cyrillic letters".
A: Yes, it is interesting, but mostly in demonstrating the informant's single-mindedness. First, it is irrelevant to ostensible influence of "Russian transliteration" on romanisation. Using a Cyrillic letter that bears no resemblance to a Roman letter is just that -- openly using a Cyrillic letter, not an influence on choice of Roman letters. Second, during the transition period from 1955 to 1958 (progressively abandoning facets of Beila in favour of eventually finalised Pinyin) some of the four trial versions of Pinyin variously:
1) Used h everywhere that Pinyin now uses x (so, "hiwang" rather than "xiwang");
2) Anglophone-style used j ch sh for both zh ch sh and j q x;
3) Attempted to realise the goal of “一音一符” (i.e. abolition of double letters) among the consonants, and to this end used five non-Latin symbols, four of which were IPA symbols and one of which was Cyrillic. (The Cyrillic letter was ч and was used where finalised Pinyin uses j. It was chosen not because of its sound value in Russian ... still less its sound value in transcription of Chinese [ ч transcribes Pekinese ㄔ not ㄐ ], but because of, as was openly stated at the time, its resemblance to ㄐ. In other words, a symbol from the politically unacceptable Zhuyin Fuhao was let in through the back door under the pretext that it was a Cyrillic letter.) As for the four IPA symbols, they bear no resemblance whatsoever to Cyrillic letters. The informant's "one or two Cyrillic letters" either means he doesn't know IPA symbols when he sees them, or means he's simply repeating old hearsay from the unsinkable battleship Taiwan.

Beila       Pinyin      ZyFh     Cyrillic Russian transcription:

b　　　　b　　　ㄅ　　　б
p　　　　p　　　ㄆ　　　п
d　　　　d　　　ㄉ　　　д
t　　　　t　　　ㄊ　　　т
g　　　　g　　　ㄍ　　　г
k　　　　k　　　ㄎ　　　к
x　　　　h　　　ㄏ　　　х
z　　　　z　　　ㄗ　　　цз
c　　　　c　　　ㄘ　　　ц
s　　　　s　　　ㄙ　　　с
zh      zh      ㄓ      чж
ch      ch      ㄔ      ч
sh      sh      ㄕ      ш
rh       r      ㄖ      ж
g/z      j      ㄐ      цз
k/c      q      ㄑ      ц
x/s      x      ㄒ      с

some syllables:
gi/zi　　ji　　ㄐㄧ　　　цзи
z　　　　zi　　ㄗㄧ　　　цзы
zh　　　 zhi　　ㄓ　　　 чжи
rh　　　 ri　　 ㄖ　　　 жи

some finals:
i　　　　i　　　[衣]　　 и
※　　　 i　　　[资]　　 ы
※　　　 i　　　[知]　　 и
y　　　　ü　　　ㄩ　　  юи
r　　　　er　　　ㄦ　　  эр
n　　　　n　　　　　　　 нь
ng　　　 ng　　　　　　  н
※ = "none"

Further Beila and Pinyin comparisons:
Beiging        Beijing
Tianzin        Tianjin
Cingdao        Qingdao
Chungking      Chongqing
zinlai         jinlai (进来)
ginlai         jinlai (近来)
rhanxou        ranhou
Xanz           Hanzi
uja            wuya
pinjin         pinyin
iu             you
iou            you (有)
mai            mai
maai           mai (买)
zai            zai
zaai           zai (再)
hoping         heping
geming         geming
zgi            ziji
sgi            siji
zs             zisi
rcie           erqie
rz             erzi
rhz            rizi
is             yisi
iz             yizi
zhego          zhe ge
naagian fangz  na jian fangzi (哪)
zinxingzhung   jinxing zhong
uniancian      wu nian qian
ulid           wuli de (的≠得、地)
nid            ni de (的≠得、地)
zui dad        zui da de (的≠得、地)
giangde xao    jiang de hao (得≠的)
zigide gungzo  jiji de gongzuo (地≠的)

The Beila orthography/word-linkage used above is based on the 1949 republication of 《新文字与新文化运动》 by 吴玉章.Interested persons will find a wealth of historical material about Beila in the article 「拉丁化新文字」in the 语言文字 volume of 《中国大百科全书》. Details of romanisations beginning from 1605 are found in 《汉字改革概论》 by 周有光. The full syllable-by-syllable Russian Cyrillic transcription of Pekinese can be seen, among other places, at the back of 《汉俄小词典》.

And finally, a further piece of evidence against the informant's suggestion that "use of the voiced consonant graphs for unaspirated consonants" is variously Russian or perverse: Among Huizu as far east as Shandong there existed until quite recently a centuries-old system of writing Chinese in the Arabic alphabet (i.e. not a mere transcription) called 小经 or 小儿锦. In G.I. Zavyalova's description of this system -- Sino-Islamic Language Contact along the Great Silk Road: Chinese Texts Written in Arabic Script (《汉学研究》〔台北〕第1卷第一期，p.294) -- we read: "[L]etters for the Arabic voiced and unvoiced consonants are used for the Chinese unaspirated or aspirated initials respectively[.]"