uwePhillPhelps / uweAudioSegmentation

applescript to monitor dropbox, PHP interface to manage tools (FAAD/LAME/shnTool/mp3splt) which convert & segment uploaded audio
1 stars 0 forks source link

audioSegmentation

What is it? This is a component part of the UWE transcription project. It consists of applescripts, shell scripts, PHP code, and sourcecode for commandline tools for audio conversion, segmentation, and timecoding. The current version is for OSX 10.6.3, has no automatic installation procedure. Instructions for manual installation are NOT included.

The purpose of the audioSegmentation component is to:

* monitor an Apple Filesharing dropbox
* automatically prepare newly "dropped" files for segmentation
* convert and segment audio
* provide a web interface to:
    * preview timecoded skeleton-segmentation TRS XML 

It has been hacked around a bit to provide additional functionality (originally intended to be part of uweTranscriptionHub):

* providing a web interface to:
    * control the segmentation process
    * produce subtitles (in SRT format) by combining RTF/TXT segments and TRS XML timecode

Requirements This component has only been tested on OSX 10.6.3.

The application depends on /usr/bin/bash for shell scripting, also LAME 3.98.4, also modified versions (included) of mp3splt 2.2.1, libmp3splt 0.5.2, libmad 0.15.1b, and shntool 3.0.10, also textutil, Applescript, SystemEvents, ...more? you betcha!

The current version makes no attempt at robustness. It will fail miserably if dependencies are not installed exactly as on the development system.

How do I use it? The current version does NOT have an automatic installation procedure. No instructions are provided for manual installation or configuration.