The Next Voice You Hear Will…Oh, Forget It.

My plans for voice recognition software were thwarted.

As Jane Smith —  from How Publishing Really Works — pointed out in the comments section, voice recognition software is fairly voice specific.  You have to “train” it to recognize your own voice, at which point you can play your own recorded voice back to it (or speak through a microphone) and the program will recognize your own words well enough to come up with a reasonable transcription.

My problem, however, is that that’s not really what I needed.  I wanted to be able to play back an interview between two people, and have the VRS system be able to transcribe it.  That, alas, is beyond the capability of most VRS systems.

The literature for MacSpeech didn’t really make that clear — I thought it was going to be a technological wunderkind, capable of transcribing whatever I might play through it (“Revolution 9” from The Beatles might have been fun), no questions asked.  That wasn’t the case — and since I don’t work by dictating into the computer, Scribe is pretty much a useless program for me.

Unfortunately, when I called customer service at MacSpeech to see if I could get a refund on the program — since it really didn’t do what I needed it to do — they told me no dice, since the program “was working as it was supposed to.”  Rats.

So I’ve gone back to Plan B — having the conversation transcribed.  I did learn, however, that if your transcription doesn’t have to carry a standard of  “legal weight” — meaning it won’t be scrutinized in a courtroom — you can have things transcribed for a much more reasonable rate.  I’m supposed to have my transcript back soon.  I’ll let you know how they did — and if it looks good, I’ll let you know who I used.