robert@psych.UUCP (04/23/87)
We have a researcher here that wants to do an experiment as follows - Subjects will presented with 2 words. Either word can be presented centered on the screen OR audibly. After a series of such presentations the subject is given a recall test. Now the problem is that the narrator device is not very clear when used for single words. 'peach' can sound like 'teach'. 'clay' can sound like 'play'. My question is - Is there any way the quality of the narrator device can be improved ? If not, then how does one go about digitising speech and the playing it back thru the audio device ? Any examples or suggestions on the use of the narrator device for speech or speech on the Amiga in general would be appreciated. Thanx, Robert Dal Santo Phone: +61 7 377 4285 (International) (07) 377 4285 (National) Department of Psychology, ACSnet: robert@psych.uq.oz University of Queensland, ARPA: robert%psych.uq.oz@seismo.css.gov St Lucia, Brisbane, 4067 UUCP: ..!seismo!munnari!psych.uq.oz!robert AUSTRALIA. JANET: psych.uq.oz!robert@ukc
dillon@CORY.BERKELEY.EDU.UUCP (04/23/87)
Right, you can digitize speech externally, somehow get the data into your Amiga, and simply slam it out one of the Audio DMA channels. -Matt
foy@aero.ARPA (Richard Foy) (04/24/87)
In article <8704232050.AA21881@cory.Berkeley.EDU> dillon@CORY.BERKELEY.EDU (Matt Dillon) writes: > > Right, you can digitize speech externally, somehow get the data >into your Amiga, and simply slam it out one of the Audio DMA channels. > > -Matt Also there are several sound digitizers available that will digitize directly into one of the Amiga ports and allow the resulting sound sample to be saved to disk. The sounds samples can then be used in whatever way you want to as Matt suggests. Richard
roliver@nmsu.edu (Richard Oliver) (03/21/90)
I am interested in a robust english to phoneme translation capability for the Amiga. The translator.library provides a good starting point and I would like to expand on that. Questions: 1. Can the standard translator use custom exception tables? If so, how? 2. If not, does anyone have translator up and running that they would be willing to place in the public domain? Thanks.