This is my version/demo of Prof Alexander Arguelles' Shadowing technique. I only started doing this after completing several audio-linguistic courses.
The basics... listen to 1 minute segments in your target language (Spanish) and repeat immediately after the audio. You are literally speaking "over" the audio, so at first it feels almost disorientating. You have to cut up the audio yourself, and you will also need a copy of the book in Spanish and, if possible but not necessary, English.
I do it seated, with headphones. I only have the head-phones on one ear, and the other ear I have open so I can hear myself (important - otherwise you end up thinking you're speaking well when you're not. IE: people who sing with head-phones on).
I slow the audio to about 70% of speed using Windows Media Player. The reason being, the vocab in Harry Potter, at that speed, is just too fast for my level of learning.
To begin with I have the Spanish text infront of me and I literally read and shadow simultaneously. I do this with a specific section until I can "manage" to keep up with the audio easily. After that I translate the Spanish text so that I consciously understand all the words in that segment, then I begin shadowing without the text.
Every so often I record myself, like in this video, to check my pronunciation. Also I occasionally refer back to the text in Spanish in order to make sure I'm not making any consistent mistakes (such as mistaking a Que for a De or something equally simple but important.)
I shadow in 25-40 minute intervals, on average about once a day. I usually do about 3 or 4 different one minute sections. So I'll do one section, on repeat, for about 10 minutes, then another, then another.
Once I'm finally done I chuck any new vocab I've discovered which I like into my Anki deck in a sentance which I'm likely to use. One good recent example is the phrasal verb - "darse cuenta" which means "to give yourself account" literally but really means "to consider" or "to realise".
I've been using this technique for about 3 months. I've noticed a enormous improvement in my pronunciation and speaking, but so far my "everyday" comprehension has not yet gained noticably from this technique. Although on top of this, it has been a great way of "discovering" new vocab.
The "trick" to making this technique work is concentration. You need to focus on the audio, ensure that your pronoucing your words as accurately as possible (there's many "levels" of accuracy which only become clear after continued exposure), and eventually to try to "think" in the target language while also understanding the meaning (this is the bit I'm struggling with).
Usually afterwards the words of the language are buzzing through your head... which is probably a good thing.
Coffee goes well with this technique ;)