102 messages over 13 pages: << Previous 1 2 3 4 5 6 7 ... 9 ... 12 13 Next >>
Paco Senior Member Hong Kong Joined 4277 days ago 145 posts - 251 votes Speaks: Cantonese*
| Message 65 of 102 02 July 2014 at 6:03pm | IP Logged |
Before giving any judgment, it is better to state clearly my limitation: I have not had
any exposure to other recorded learning materials, and I can only speak for the variety
of Cantonese spoken in Hong Kong. As it is a native language in many places in Asia and
spoken by varying sizes of groups of people worldwide, debate about the
standard/dialect/accent" thing never stops.
The book may be well worth having, and we will have to wait and see, but the audio
leaves a lot to be desired. The best voice is the clear, bright female one - actually I
quite like it - though using exaggerated actor's intonation and with wrong
pronunciation at times, the voice is clear enough for producing recordings and she
speaks with spirit and without "lazy sounds*"; I would have recommended complementing
the book with the audio had you chosen to use the course and everything been recorded
by her. As for the other (3?) voices, I would avoid them all. They lack appropriate
intonation, pronunciation, and even control of tones in some cases. Based on those 3
clips available (and them only), I would advise against learning with the recordings.
--------------------
As a side note, I would like to make a guess. The story might be like this: Not only
does Assimil think these people are native speakers, they believe they are, too, as
offsprings of Cantonese people. Perhaps in that sense they count, but they have not had
enough exposure to "correct" Cantonese and have been influenced by French (No one
pronounces 堂 (lesson) without some kind of French r sound). But this is just a wild
guess.
*I do not know the proper expression for 懶音 (which literally means lazy sound) in
English. Wikitionary tried but I am afraid the entry is inaccurate, because we
think the phenomenon of the loss of nasal consonants (e.g. "n" replaced by "l") counts
as 懶音.
Edited by Paco on 02 July 2014 at 6:16pm
8 persons have voted this message useful
|
newyorkeric Diglot Moderator Singapore Joined 6379 days ago 1598 posts - 2174 votes Speaks: English*, Italian Studies: Mandarin, Malay Personal Language Map
| Message 66 of 102 03 July 2014 at 3:40am | IP Logged |
Thanks, Paco. It's a shame because a lot of us have been looking forward to the course.
Anyone have a listen to the Khmer samples yet? Any opinions? This is also a course on my radar.
1 person has voted this message useful
| vermillon Triglot Senior Member United Kingdom Joined 4678 days ago 602 posts - 1042 votes Speaks: French*, EnglishC2, Mandarin Studies: Japanese, German
| Message 67 of 102 03 July 2014 at 8:19am | IP Logged |
Antanas wrote:
vermillon wrote:
QiuJP wrote:
I think Assimil has released a new Japanese course for French speakers: Le Japonais |
|
|
It is only the reunion of the old Japonais sans peine, volumes 1 & 2, in a single book. It is now available as a Super Pack, but that's about the only good thing about this new version, so for anyone already having the old versions, there is no need to look after this new release. |
|
|
You are both right and wrong. They say in their blog that recordings are 20% new in comparison to the older ones. And that only what concerns dialogues. Also, they recorded texts of 14 revision lessons:
3. Parlez-nous des enregistrements. Quelles est la proportion d’enregistrements nouveaux dans cette nouvelle édition ?
C.G. : Pour les dialogues la proportion est de 20%. Il s’agit soit de dialogues entièrement nouveaux soit de dialogues modifiés pour des raisons de vocabulaire. Et il ne faut surtout pas oublier les enregistrements de dialogues de révision (qui n’existaient pas dans la précédente édition), à la fin de chaque leçon de révision : soit 14 enregistrements de 10 phrases chacun. |
|
|
After a few weeks, I can comment on this again: 20% must include the 14 revision lesson recordings: having reached L36 I have found only 2 lessons which have been modified from the old edition. These modifications are extremely minimal: they changed the name of a place name (there's a bigger tower in Tokyo now..) and changed the word "home drama" to "drama" (TV series).
I'll even go further: the revision lesson recordings are not recordings: they're text-to-speech generated. They've used a quite good TTS engine, but several things make it noticeable:
-25 years later, the actors have exactly the same voice: this seems unlikely. (but isn't my strongest evidence)
-completely unnatural pace of the sentences: much faster than in the rest of the dialogues and with longer pauses after grammatical particles. I have trouble following these sentences, while the rest of the book is fine to me.
-text-to-speech artefacts! The software they used is pretty good, but nonetheless in several sentences you hear some artefacts due to poor sound concatenation.
3 persons have voted this message useful
| Arnaud25 Diglot Senior Member France Joined 3842 days ago 129 posts - 235 votes Speaks: French*, English Studies: Russian
| Message 68 of 102 11 July 2014 at 7:42pm | IP Logged |
vermillon wrote:
After a few weeks, I can comment on this again: 20% must include the 14 revision lesson recordings: having reached L36 I have found only 2 lessons which have been modified from the old edition. These modifications are extremely minimal: they changed the name of a place name (there's a bigger tower in Tokyo now..) and changed the word "home drama" to "drama" (TV series).
I'll even go further: the revision lesson recordings are not recordings: they're text-to-speech generated. They've used a quite good TTS engine, but several things make it noticeable:
-25 years later, the actors have exactly the same voice: this seems unlikely. (but isn't my strongest evidence)
-completely unnatural pace of the sentences: much faster than in the rest of the dialogues and with longer pauses after grammatical particles. I have trouble following these sentences, while the rest of the book is fine to me.
-text-to-speech artefacts! The software they used is pretty good, but nonetheless in several sentences you hear some artefacts due to poor sound concatenation. |
|
|
I was intrigued by your comment, so I've gone to the library this afternoon to "borrow" the audio.
For me the lessons of revision are read by human beings. I've listened to the lesson #7 and I can understand it correctly. The voices are different than in the other lessons, so for me Assimil has engaged other actors to read them. It's true that the lessons of revision sound like less "articulated" than the dialogs.
It could be TTS in the "modified" lessons, but it's difficult to be completly positive on that too: comparing the old and new audio track of the lesson #6 for exemple: the voices of the girls are different (but very similar), the voice of the guy is the same but the intonations, the です ("dess"/desu), the "L/R" sound are not similar in the old and new track: they have not replaced the different words, all the track is slightly different. It could have been read again by the same person (the voice is one of the thing that change the less along life: once you have your adult voice, if you don't smoke and drink too much, it's almost always the same until you're very old)
But it's true that sometimes you hear the "concatenation" effect, I've heard it too.
For me, it's not TTS, or if it's TTS it's a real good one.
After that, I can't tell you if it's worth the money: I'm a good client of libraries (I've just bought the book and I'm slowly ploughing my way through it)
Last note: In the first lessons, the dialogs are read rapidly and then slowly: it was the contrary on the old audio: the lessons started slowly and were repeated a second time more quickly.
Edited to correct typos, sorry for my poor English.
Edited by Arnaud25 on 12 July 2014 at 9:20am
1 person has voted this message useful
| vermillon Triglot Senior Member United Kingdom Joined 4678 days ago 602 posts - 1042 votes Speaks: French*, EnglishC2, Mandarin Studies: Japanese, German
| Message 69 of 102 12 July 2014 at 10:39pm | IP Logged |
Well, I'm not saying it's not worth the money: I bought it and love it.
Without making it an argument of authority, I have studied and researched TTS, so I've had quite a lot of exposure to what is TTS and what is not: if you still have it, listen to, for instance, the second sentence of lesson 35 (海で泳いでいた時、太陽が水 平線から出てくることを見ることができまし た。すばらしかったですよ。): you will clearly hear that:
1) it's robotic, the delivery is fast and monotonous.
2) there are jumps in the sound: that is because at some point the audio segments that the TTS software chose didn't match particularly well. One instance is around the "見ること"
Of course, now we're in 2014, so TTS is much better than it was some years ago but it's still TTS. I mention it because even though I love Assimil, I don't like it when they publish an interview for marketing reasons and this interview is clearly exaggerating the efforts put into that new version (minimal, imho).
Edited by vermillon on 12 July 2014 at 10:42pm
1 person has voted this message useful
| Arnaud25 Diglot Senior Member France Joined 3842 days ago 129 posts - 235 votes Speaks: French*, English Studies: Russian
| Message 70 of 102 12 July 2014 at 11:46pm | IP Logged |
vermillon wrote:
If you still have it, listen to, for instance, the second sentence of lesson 35 (海で泳いでいた時、太陽が水 平線から出てくることを見ることができまし た。すばらしかったですよ。): you will clearly hear that:
1) it's robotic, the delivery is fast and monotonous.
2) there are jumps in the sound: that is because at some point the audio segments that the TTS software chose didn't match particularly well. One instance is around the "見ること"
Of course, now we're in 2014, so TTS is much better than it was some years ago but it's still TTS. I mention it because even though I love Assimil, I don't like it when they publish an interview for marketing reasons and this interview is clearly exaggerating the efforts put into that new version (minimal, imho). |
|
|
You're right, unfortunately. I had listened only to the beginning of the book, as I'm only at the lesson #5 today. It's less evident at the beginning of the book than at the lesson #35, probably because the sentences are shorter.
The lesson #35 is terrible, I hope it's an isolated accident.
1 person has voted this message useful
| Mutant Groupie United States Joined 3911 days ago 45 posts - 60 votes Speaks: English* Studies: French, German
| Message 71 of 102 06 August 2014 at 1:50pm | IP Logged |
newyorkeric wrote:
Thanks, Paco. It's a shame because a lot of us have been looking forward to the course.
Anyone have a listen to the Khmer samples yet? Any opinions? This is also a course on my radar. |
|
|
My Cambodian-American friend, who can speak and understand a lot of Khmer, said that Assimil's recordings on Soundcloud sound very good, but also very formal. She said they remind her of the way her mother would speak, not someone around our age (I'm 29). So I'm counting on her to teach me more "colloquial" speech. I played lessons 1 and 100 for her and she was able to keep up and translate it pretty well. She didn't really say whether or not she thought the voices sounded unnatural.
2 persons have voted this message useful
| Mutant Groupie United States Joined 3911 days ago 45 posts - 60 votes Speaks: English* Studies: French, German
| Message 72 of 102 14 August 2014 at 12:25pm | IP Logged |
I ordered Assimil Le Khmer this morning. I feel confident my French is good enough to use it, although it may take me a little longer. After I have really gotten into it, I will post a short review and I will also report on what my native-speaker friend thinks about the course.
I am very excited, I have been waiting for this for a long time!
1 person has voted this message useful
|
You cannot post new topics in this forum - You cannot reply to topics in this forum - You cannot delete your posts in this forum You cannot edit your posts in this forum - You cannot create polls in this forum - You cannot vote in polls in this forum
This page was generated in 0.6406 seconds.
DHTML Menu By Milonic JavaScript
|