harmo@cc.helsinki.fi (Timo Harmo, Fac. of Soc.Sci, U of Helsinki) (04/07/90)
Does anybody know of systems for voice-databases connected to hypercard (or something similar). What we woud need is a database of several hours of speech-samples that could be accessed from the corresponding texts (litterations of the samples). I think it woud require voice-compression, saving them in files in a worm-drive (?), some simple dbms, and an interface (xcmnds) to hypercard. If there isn't anything like that for hypercard, do you know any newsgroups where I could ask about other ways of doing this kind of databases (comp.databases.voice, sci.speech.analysis :-) -Timo Harmo
taylorj@yvax.byu.edu (04/13/90)
The main problem is that you want several hours of speech. If you stored the sound at 6:1 compression (generally adequate for speech targeted for a native listener), it would take about 13 megabytes per hour of speech. So you'll either need a hard disk or CD-ROM. The rest of it (indexing, accessing the sounds, etc.) should be pretty easy from HyperCard. Jim Taylor Microcomputer Support for Curriculum Brigham Young University taylorj@yvax.byu.edu