Discuss and learn French: French vocabulary, French grammar, French culture etc.
Could you be more specific?
I don't think that Google would really bother with pronunciation of liaisons and moreover as a rule never rely on Google translations!
Incidentally few people know that all translations are only based on English language. For example let's say you want to translate German into French, Google will first translate into English and then into French so whatever the language the translation will always be influenced by English syntax vocabulary and idioms.
If you use Google Translate frequently, as I do, you will see that sometimes the last consonant is pronounced and sometimes not, as if it understands liaisons. In fact, when I double check on these it often gets the liaisons right. I'm just not sure how often they are right. I was wondering if anyone knew about this. Vedas, if you think that Google Translate never pronounces liaisons, then you don't really know. I would appreciate commentary from a knowledgeable source if there is one out there.
When you listen to Google Translations is it a machine talking or is it a native speaker?
If it is a machine you are wasting your time in my opinion (regarding pronunciation ,that is).
It is a machine. However, there are real humans constantly improving the translations. It is a very helpful tool though it does make mistakes.
I didn't say"never"
but you sem so full of yourself that maybe the best solution would be to ask the machine itself!
You don't seem to get along with humans very well...
Hi Marcus --
Like any text-to-speech system, the system used by Google Translate uses some simple statistics to decide whether to pronounce the liaison consonant. So in general, it will probably pronounce the liaison correctly in very basic cases where there's little chance of not pronouncing it, e.g. between pronoun and verb or between an adjective and corresponding noun.
In other cases where there is variation anyway depending on speaking style etc, then results will be more variable, e.g. the system tends to "over-estimate" the likelihood of liaison in some cases where it's fairly rare in speech generally.
But I'd see it like this: it gives you a good "first stab" at whether or not there'll be liaison in a fairly formal, emphatic style of reading.
a more fluent speaker ? il faudra apprendre à être poli pour commencer si vous voulez une réponse...
I meant a more fluent speaker than myself. Vedas, I do apologize for sounding curt in my earlier comment. I can see you have been quite helpful to others on this site. I will always appreciate anything you have to offer.