When it comes to fluency, you might want to look at some of Prof.Cem Alptekin's work (which can be found on Researchgate) I have been teaching IELTS classes in Vancouver, and have one of your students here (I mean from Hacettepe ) Prior this position I had been running Power Speaking courses which had consisted of speaking techniques, elocution and accent reduction. I would like to summarise my personal findings. The difficulty in pronouncing English differs depending on the sound system in mother tongue. Turkish people, for example might have problems with 'th' sound which partially might depend on inappropriateness of showing tongue (cultural); as well as on the unfamiliarity of tongue muscles with that move. Chinese, having one syllable word system, might find it difficult to pronounce the end consonants; Japanese, Koreans, and Mexicans might have some problems with minimal pairs such as l/r, p/f, b/v respectively.. There are some studies about the effects of L1 on L2 in Google Scholar. Here are some : http://www.sciencedirect.com/science/article/pii/S0095447096900406
I would like to add another observation, which again would not be based on research but rather experience. Since you are dealing with French; what I realized is, the "throat sounds" like "R" in French (another more rolled form in German) is also difficult for some speakers to produce, and vice and versa.Also, lexical differences might play a role. English, being a non-phonetic language, makes it difficult for speakers of phonetic languages due to its silent or 'differently sounding' letters (vowel sounds/vowel letters difference). Also,seeing written words, it is difficult to say "hotel" to a Turkish or German speaker without pronouncing the "h" sound. Let's look at these links;
I remember having a sore throat to get used to German "R" and also, my tongue was a little hurting while trying to get used to "tongue rolled back" "R" like in North American English. I was taking pronunciation too seriously since I was going to teach it too. Despite the fact that in academia, especially some believers of post structuralism doubt the necessity of "native-like" speaking, I have been a firm believer of the "phonetic proficiency/competency". Nevertheless, we might consider the 'capability of learners" in producing the "foreign sounds" and not push it too hard for sure, as long as their discourse is intelligible"
I have just come upon your conversation about fluency. A couple of years ago, the research group in the Digital Media Centre here at The Dublin Institute of Technology in Ireland did a lot of research on fluency (see our publications on ResearchGate on our projects FluenCi and Fluent). initially, we started looking at the mis/pronunciation of phonemes after reading Jenifer Jenkins work on English as an International Language (Jenkins, J. 2001. ‘The Phonology of English as an International Language’. Oxford: Oxford University Press). Jenkins showed that only mispronunciation of certain phonemes inhibit intelligibility, wheras others don't. From there we started looking at the difference between pronunciation and fluency, or perceived fluency. We became very interested in the work of Alison Wray (Wray, A. 2002. Formulaic Language and the Lexicon. Cambridge: Cambridge University Press. and Wray, A. 2004. Formulaic language learning on television. In: Formulaic Sequences, N. Schmitt (ed.). Amsterdam: John Benjamins Publishing Company.), who discovered that what actually makes a person sound fluent is the use of frequently occurring sequences, or as she calls them, formulaic sequences. Such multi-word sequences are processed very efficiently in the brain and can therefore be spoken faster and are also 'unpacked' faster by the listener. As such sequences can therfore roll of the tongue as a unit, it makes the speaker sound more fluent - added to that is, in my opinion, the added bonus of the speaker possibly sounding as if they are part of the cultural target group. I really noticed this with some of my non-Irish students, who started using the word 'like' or even the phrase 'like, you know like' in their informal conversations which were otherwise peppered with pronunciation mistakes. It somehow made them sound more natural, despite the mispronunciations.
Another interesting angle is provided by John Field, who looks at the importance of word stress and syllable stress on listening, intelligibility and pronunciation (Field, J. 2003. Psycholinguistics. A resource book for students. London: Routledge.) and particularly (Field, J. 2000. ‘Not waving but drowning: a reply to ‘Tony Ridgway’. ELT Journal 54/2 (pp.186-95). and Field, J. 2003. Promoting Perception: Lexical Segmentation in L2 Listening. ELT Journal Volume 57/4 October (pp. 325-333).