Register  Login  Active Topics  Maps  

TIP: Automatic audio-splitting, Audacity

 Language Learning Forum : Links & Internet Resources Post Reply
24 messages over 3 pages: 1 2 3  Next >>
ryuukohito
Bilingual Diglot
Groupie
Malaysia
Joined 5986 days ago

89 posts - 98 votes 
Speaks: EnglishC2*, Malay*
Studies: French, Japanese

 
 Message 1 of 24
09 January 2008 at 10:24pm | IP Logged 
TIP: How to automate the audio-splitting process to help with shadowing

Here's something I've discovered whilst using Audacity. I would like to share it with my language-learning brethrens who also use Audacity for shadowing purposes:

1. Load the audiobook mp3/wave file in Audacity.

2. Select all (keyboard shortcut: CTRL-A) or a portion of the audiobook section.

3. Click Analyze, and select Silence Finder. (This feature will analyze portions of audio demarcated by silence, and it will label them accordingly. At this point, it is immensely helpful if the audiofile you are using has only the narrator/storyteller speaking at all times, with no accompanying background music.)

4. I have found a useful generic setting for almost all of my audio files to be this:

Silence Level: 15 (dB)
Minimum silence duration: 0.5 (seconds)
Place label: 0.30 (seconds)

(Adjust accordingly. A brief explanation of the above things: The 'silence level' determines the minimum 'volume' of silence a portion of audio has to be at before it is considered 'silent', and is made as a marking point. Hence, if there is noise, or background music, in the audio, consider raising this value. The 'minimum silence duration' is the minimum length between two points in an audio file where there is silence. Hence, if the speech going back and forth, or narration between two sentences, is too fast, consider decreasing this value. The 'place label' determines how much space, in seconds, is placed in front of each labelled audio portion.)

4. If all goes well, you will now see a new 'track' appear. Check if everything is right, by pressing the CTRL key and scrolling with your mouse scroll-button to zoom in.

5. If the audio portions have all been correctly 'labelled' (and split), click File -> Export Multiple. Select your export format, and adjust the export directory and naming settings, then click Export.

6. What happens is that the audio will then be split accordingly into many various sections, usually each a chunk of 1-5 seconds long. (All silence in between have been removed, so you get chunks of pure audio data only.)

You can then load these mp3s in your favourite media player (Winamp, Foobar, Windows Media Player, VLC, et cetera, although I prefer the last because it makes it really easy for me to loop one particular chunk endlessly) and choose to focus on and repeat/shadow only the particular chunk of audio you fancy at that time, to practice intonation, pitch, melody, and whatever other colourations of speech as need be. (This avoids you having to waste time selecting manually the portion of speech you need to practice in Audacity.)

This has helped me a lot, the chunk-approach (as I call it). I even carry around these split-up mp3s with me everywhere for practice during those spare moments, and they really are a nice way of practicing things, because you're not overloading your brain with too much audio data at one time. (Small chunks work best!)

I hope this tip will help you. Let me know if it does.


Edited by ryuukohito on 10 January 2008 at 5:14am

6 persons have voted this message useful



Sulis
Groupie
United Kingdom
Joined 6201 days ago

60 posts - 66 votes 
Speaks: English*
Studies: French

 
 Message 2 of 24
09 January 2008 at 11:18pm | IP Logged 
Great tip - thank you! I'll give it a try.
1 person has voted this message useful



maxb
Diglot
Senior Member
Sweden
Joined 6933 days ago

536 posts - 589 votes 
7 sounds
Speaks: Swedish*, English
Studies: Mandarin

 
 Message 3 of 24
10 January 2008 at 2:43am | IP Logged 
Thank you very much. I have been doing chunking for almost a year now. Putting them into supermemo to memorize them better. This function is just what I need. Saves me the trouble of splitting the audio manually.
1 person has voted this message useful



delectric
Diglot
Senior Member
China
Joined 6931 days ago

608 posts - 733 votes 
Speaks: English*, Mandarin
Studies: German

 
 Message 4 of 24
10 January 2008 at 10:49am | IP Logged 
Yeah i've been doing the same with supermemo too. Anyone got any tips for Cubase SX?
1 person has voted this message useful



Marj
Senior Member
United States
Joined 6316 days ago

257 posts - 283 votes 
Speaks: English*
Studies: Spanish, Mandarin, French

 
 Message 5 of 24
10 January 2008 at 12:25pm | IP Logged 
Thanks for the tip. I especially appreciate that you break down the steps, so it is clear exactly how it can be done if one isn't very computer savvy.
1 person has voted this message useful



matematikniels
Tetraglot
Groupie
Denmark
Joined 6002 days ago

78 posts - 84 votes 
Speaks: Danish*, English, German, Swedish
Studies: Russian, Spanish

 
 Message 6 of 24
10 January 2008 at 5:21pm | IP Logged 
I found the multipurpose audio file editors like Audacity to slow for this job, so I prefer to use a dedicated program. A few specialized Windows programs for this task:

Open Source Audiobook Cutter and freeware Slice.

I use Slice because I got used to it. It lets you split multiple files at the same time.

1 person has voted this message useful



ryuukohito
Bilingual Diglot
Groupie
Malaysia
Joined 5986 days ago

89 posts - 98 votes 
Speaks: EnglishC2*, Malay*
Studies: French, Japanese

 
 Message 7 of 24
11 January 2008 at 6:56am | IP Logged 
Marj wrote:
Thanks for the tip. I especially appreciate that you break down the steps, so it is clear exactly how it can be done if one isn't very computer savvy.


I'm glad you found it helpful, Marj!

And, to matematikniels, I was not aware that there was such a software. I had done a run through Google but had not encountered it before this. Thank you!

(NOTE: I have tried AudiobookCutter, and it is very good. However, it seems the minimum size one can split an audiobook to is 15 seconds. In comparison, Audacity allows one to split into even smaller chunks of 1-3 seconds. I humbly believe that a chunk of 1-3 seconds allows one to shadow a sentence with greater ease. Sometimes, shadowing a line of 15 seconds can be quite difficult, so that is why I would still suggest the Audacity chunk-making method. Unfortunately, my only authority for suggesting that smaller chunk sizes are better is the John Petrucci chunk-method I presented in another post of mine. I hope it suffices to convince you.)

Edited by ryuukohito on 11 January 2008 at 7:07am

1 person has voted this message useful



LpMagilicutty
Newbie
United States
Joined 5937 days ago

24 posts - 26 votes
Speaks: English*
Studies: Spanish, Mandarin, Tagalog, German, Italian, French

 
 Message 8 of 24
13 January 2008 at 1:18am | IP Logged 
ryuukohito,

Brilliant!
It was taking me forever trying to do it manually. Thanks SO MUCH for sharing.


1 person has voted this message useful



This discussion contains 24 messages over 3 pages: 2 3  Next >>


Post ReplyPost New Topic Printable version Printable version

You cannot post new topics in this forum - You cannot reply to topics in this forum - You cannot delete your posts in this forum
You cannot edit your posts in this forum - You cannot create polls in this forum - You cannot vote in polls in this forum


This page was generated in 0.3750 seconds.


DHTML Menu By Milonic JavaScript
Copyright 2024 FX Micheloud - All rights reserved
No part of this website may be copied by any means without my written authorization.