Improving pronunciation and prosody

Here's one I haven't seen anyone mention explicitly, I just call it Modeling:

• Open up Audacity, load a piece of audio, select a phrase and hit space to listen to it a few times.
• Next, hit play again and try repeating the phrase AT THE SAME TIME using the same intonation, stress and pronunciation. Do that a few times until you reach your best.
• Whenever a chunk of speech (could be one sound, could be something longer) is not clear to your ear, zoom in, select it and repeat it.

You progress phrase by phrase (NOT sentence by sentence.)

The main distinction here, compared to other uses of Audacity, may be that you repeat things simultaneously, the goal being to SYNCHRONIZE yourself with the original. You might not get all of the sounds right but I believe ANYONE can get the rhythm right this way... and it's already quite something in terms of getting closer to native speech (think of it as a grid: if you get the rhythm right, it'll be easier for the rest to fall into place. )

(Note: since I first experimented with this, I've learned that you could also just use Shift+space to repeat the selected audio indefinitely but I like tapping play myself. Makes me feel in control I guess. I would often tap on my desk at the same time, as described earlier in this thread, to mark the stress and add something physical to the whole concept of rhythm.)

I had the misfortune of learning most of my English from text, like described in the original post, and it does take an awful lot of work to get rid of bad habits. A phonology class helped me fix my pronunciation by chance but only the above helped me improve my prosody significantly as far as I can tell.
I also tried posting stuff on SoundCloud to try and get feedback but without much success. I probably would have needed a partner in crime to make this feedback part work!)

---

Here are my two cents about a few things related to this thread:

- I don't think Pimsleur's talent is in teaching phonetics but still do think it makes wonder for one's spoken language. Because it's primarily audio-based and it does not require any text support, it does a great job of helping people focus on the spoken language (duh! that's where it should all start! ). Also, it helps them uninhibit themselves from speaking in a foreign language. (All in opposition to classroom instruction and other methods.)

- Wikipedia does a good job of explaining how to articulate sounds in a foreign language. Just look for the "IPA chart" page for your target language there. I'm not claiming it's perfect but it's clearly a ton more insightful than the average language-learning book (given such information is usually no where to be found!)

- Overall, I'm dumbfounded that IPA has not spread like wildfire. It's the way of the future for anyone who wants to either develop a good pronunciation (and accent, if intonation and stress information are provided) or learn several languages.

- Learning phonetics is time consuming but it pays dividends to do so early, rather than late (provided, of course, that one does not end up giving up on the language altogether because one thinks things go too slowly!) A way to reconcile phonetics with the rest of language instruction might be to teach people to read with good pronunciation early on. I believe that's how Luca Lampariello goes about it. The benefit is then that you may enjoy learning more vocab faster (thanks to text) while keeping up with a good pronunciation. Personally, after wasting so much energy fixing my English pronunciation and accent, I've become quite paranoid about phonetics and I am learning Russian really slow because I'm being super careful (too careful?) with my pronunciation.

"Modeling" sounds exactly like the "chorusing" method we were discussing, at least my interpretation of how to do it without a teacher: listening to a phrase lots of times then speaking over it and trying to mimic it lots of times. As I say, I've found it helpful to an extent and was unsure whether it was only to this extent because I hadn't done enough of it or because of its limitations as a method.

I agree with the rest, especially about IPA. The only way I was able to learn to pronounce all the French sounds correctly was by learning about the symbols and the tongue positions for the vowels and all the rest. No amount of listening or non-scientific "A as in father"-style descriptions could have helped me understand that.