The ALLF recipe book

[Henkkles]

So I had this idea: we have lots of resourceful folks here at the forum who have invented all kinds of ways to extract language competence out of a myriad resources, so why not make a thread where we can bring these methods together and discuss them for the betterment of our repertoires!

Here, I'll start:

The ultimate listening comprehension challenge à la Henkkles

Targeted skill: listening comprehension, accent

Target proficiency: upper beginner to upper intermediate

What you need: A notebook, TL audio, transcript and translation. Can be anything, GLOSS-lessons are my favorite for this.

How to prepare the material: Open the audio in your favorite DAW such as Audacity and select around one minute of audio (~10 sentences) or maybe a bit more.

Execution:

Phase 1:
Step 1. Listen to your snippet through, write down in your notebook an approximate percentage of how much you could understand.
Step 2. Transcribe the sentences one by one having each sentence looping constantly for as long as it takes you to transcribe it. If you can't hear something, check against the transcript and try to do it again.
Step 3. After you've transcribed each sentence, translate them one by one as directly as possible making note of all new vocabulary. Use the translation as an aide if necessary.

Phase 2:
Step 1. Loop the sentences one by one trying to pronounce them on top of the recording. Do this as many times as you need for it be nearly effortless. (20-50 reps each maybe depending on your level)
Step 2. Listen to the entire snippet again, write down a new comprehension percentage and compare. You should see a huge improvement.
Step 3. If your new comprehension level was below 100%, identify the sentences that caused you trouble and repeat phase 1 for those particular sentences (with fewer reps).

Repeat until comprehension is 100%. You can also make flashcards of all new vocab you encounter to help you retain them.

This method combines almost everything I know about what works, and it makes a small amount of material last AGES; it converts one minute of audio into fifty minutes of study time (on average).

[Tomás]

For pronunciation training, I personally want to emulate the most beautiful voices I can find. I am building a Spanish pronunciation trainer for myself that anyone else is welcome to use:

https://dl.dropboxusercontent.com/u/2381631/pronunciation%20trainer2/pronunciation_trainer.html

The first two sentences are from Tatoeba. The rest are read by two male actors with stunning voices. It's a work in progress. I will be adding some clips read by Jose Duarte, another great voice. I will also be adding some clips from FSI Basic where they talk very fast and slur words together.

[Henkkles]

For pronunciation training, I personally want to emulate the most beautiful voices I can find. I am building a Spanish pronunciation trainer for myself that anyone else is welcome to use:

https://dl.dropboxusercontent.com/u/2381631/pronunciation%20trainer2/pronunciation_trainer.html

The first two sentences are from Tatoeba. The rest are read by two male actors with stunning voices. It's a work in progress. I will be adding some clips read by Jose Duarte, another great voice. I will also be adding some clips from FSI Basic where they talk very fast and slur words together.

Hey, it looks really cool, could you format it in the way I provided in the OP, namely listing these.

Targeted skill: (ie. pronunciation)

Target proficiency:

What you need:

How to prepare the material:

Execution:

[Henkkles]

Time to revive this thread.

The "Glossika schedules have too many moving parts help" man's Glossika schedule in two parts

Parte A

1. Open up a GMS-C file in Audacity or WorkAudioBook
2. Loop each sentence until transcribed to perfection, do ten sentences a day
(3. Write down new words and expressions to be imported into ANKI)

Parte B

1. Open up a GSR file in a media player
2. After hearing L1 prompt, try to produce the TL sentence in the gap
(3. If you're on a self-flagellative mood, you can write down all sentences that you fail to produce)

An alternative B-partition to that routine:

Parte B-alternative

1. After transcription, loop the sentence and do Olle Kjellin-style shadowing on it until you can pronounce it without tripping up