Six years ago (I can’t believe it’s been that long!) I made a video where I first introduced the idea of chunking in language learning.
In that video, I shared a very basic strategy for looping audio chunks using the free/Open Source audio editor, Audacity.
People seemed pretty eager to find out more about how I use the audio tool in my learning (and I STILL get questions about it today) but like a lot of things, I moved on and never talked about it again!
Well… I still use Audacity regularly in my learning.
So today I’m sharing another of my favorite Audacity tips to help you learn languages more efficiently.
Step 1: Download and install Audacity (obviously)
I didn’t mention this in my video but I’ll state the obvious here.
Audacity is free and Open Source, and is available on pretty much every operating system (Windows, Mac, Linux, BSD). I’m using a BSD version in the video which may be a few versions behind the one you get but they all do the same thing.
Download your version here.
For Linux and BSD users, it’ll depend on what package manager your OS or distribution uses (Audacity may even be included). I’m sure I don’t need to explain it but just in case you need it:
Ubuntu/Mint: sudo apt-get install audacity
Arch/Manjaro: sudo pacman -S audacity
FreeBSD: sudo pkg install audacity
OpenBSD: doas pkg_add audacity
Step 2: Import your language audio track (taken from a course, podcast, TV show, etc.)
I’m going to assume that whatever you use, you have the rights to (I shared other tips of downloading video material here).
The sample I used in this video is an Egyptian Arabic – Beginner lesson on ordering food from TalkInArabic.com (ideal because it’s spoken slower with pauses to make selecting sections easier).
If you have a video and want to extract the audio, use FFmpeg (available on all operating systems as well and it’s also Open Source).
Use this command to create an .MP3 file from any video:
ffmpeg -i video.avi -q:a 0 -map a audio.mp3
This will take a video (change ‘video.avi’ to whatever your video file is called) and create an .MP3 file called ‘audio.mp3’.
Open it with Audacity or use File > Import > Audio.
Step 3: Select the sections of the track you want to focus on and create labels
If it’s a large audio file, I recommend trimming it down and focusing on a very short part.
I always emphasize high repetition of a very small amount of language material as the most effective way to learn. If you try to learn too much, you’ll learn nothing.
When you’ve selected a section of the audio track, hit CTRL+B.
This will create a label track and insert a label at the selection point.
Do this as many or few times as you want to (I recommend not doing it to the point where you have a page full of labels — it gets too untidy).
Step 4 (optional): Export the labels to a text file in order to create subtitles (or just to save them)
Once you’ve finished making your Label track, go up to File > Export > Export Labels… and save it as a text file.
Now you can exit Audacity and edit the text file using a text editor of choice.
The syntax for a subtitle file (.SRT) looks like this:
1 (sequential numbers starting with 1 – so the next one will be 2 and so on)
00:00:00,000 --> 00:00:00,000 (hour:minute:second,millisecond)
label goes here (the label you created)
The exported text file won’t look like this, however.
It will need to be edited accordingly (use the template I’ve provided for guidance).
Did you find this language learning screencast useful or interesting?
I’m still trying to gauge people’s interest in this kind of content.
It’s a little on the technical side of things I know and far from what I’ve typically shared in the past but I’m testing it out.
Can you leave a comment below or on my YouTube video and let me know?
It’ll help me decide if I continue making more or not.
Also, if you like this style of learning (high-repetition, natural audio) but prefer something automated, then read my Glossika review. I talked about it extensively and interviewed its founder.