Voice Compression

This essay is about a suggested student project in Java programming. This essay gives a rough overview of how it might work. It does not describe an actual complete program. I have no source, object, specifications, file layouts or anything else useful to implementing this project. Everything I have to say to help you with this project is written below. I am not prepared to help you implement it; I have too many other projects of my own.

I do contract work for a living, which could include writing a program such as this. However, I don’t do people’s homework for them. That just robs them of an education.

You have my full permission to implement this project any way you please.

The intent of this project is to come up with a very low cost, low-bandwidth Internet radio. There is another project with a similar goal, Internet Radio that describes a BitTorrent-like protocol for low bandwidth broadcasting. This project focuses on how to compress the broadcast itself by creating a sort of MIDI for voice.

It works like this. Low budget Internet radio content is mostly one person talking or a small group of people talking to each other. Each gets his own noise-cancelling microphone. You analyse the speech and convert it into phonemes. What you broadcast is then a sort of phonetic text transcript of what they said, augmented by volume, and emphasis. Each speaker has a profile, much like the one Dragon Naturally Speaking compiles of how the speaker pronounces various phonemes. The phonemes don’t have to be a standard universal set, just a set of typical noises a given speaker makes in speaking. At the receiving end, the speech is reconstructed gluing together the phonemes with smoothing.

A primitive version of this would sample each word used in the transcript and create a model where each word was treated as a “phoneme”. Even something that primitive would still result in huge compression. The problem is coming up with sufficiently good voice quality that people would put up with it to listen to tiny political radio stations with no budget.

You have already seen technology similar to this in political parody.

super compressor
The George Bush speech impersonator

	Please email your feedback for publication, errors, omissions, broken/redirected link reports and suggestions to improve this page to Roedy Green :
	Canadian Mind Products
	mindprod.com IP:[65.110.21.43]
	Your face IP:[38.103.63.61]
	You are visitor number 1,502.
	You can get a fresh copy of this page from:	or possibly from your local J: drive (Java virtual drive/mindprod.com website mirror)
	http://mindprod.com/project/voicecompression.html	J:\mindprod\project\voicecompression.html