Alternative diagnosis that would qualify me for Team Gleason?

Status
Not open for further replies.
a footnote to voice banking - Tom used ModelTalker on a pc, had bought new microphone to get good quality recording and the whole process was interrupted and came to an end bc ModelTalker recording did not work any longer. According to their support is was a browser update that made the whole process not work any longer. Tom tried a few times later - hoping their programmers would catch up and eventually gave up as his voice had deteriorated too much.

He used whatever he had already recorded (about 60% of the sentences) to produce a 'voice' but we quickly figured out that it was quite some fun to use the voices that came with his operating system and ended up not using the recorded voice.
Much luck on your endeavor!
I also recently heard a report on NPR about a company in Eastern Europe (?) that has developed really good technology for imitating voices and while listening I thought that it would be of interest to the ALS community - possibly too expensive... I cannot remember the name of the company/software - it might have been "Speechmorphing" (I did a google search on NPR and generating speech - not sure whether the result and the piece I listened to about 1-3 months ago are the same - but maybe someone else has listened to this piece and remembers more?
 
With Audacity or any similar app, just make sure you specify the output format as high-quality WAVs and test multiple setups. For example, in some comp setups, sometimes an internal mic yields better sound than an external one, which is counterintuitive. And test with each of your comps/devices. It's not just the sound, it's the timbre of your voice that you want to come through.

You are just recording the word/phrase/sentence list of whatever provider you select as individual WAV files. You can record with lots of apps, like Audio Hijack, GarageBand, whatever you have that can record in high quality. Probably even a simple voice recorder. The easier the process integrates with the rest of what you do, the more likely you'll make it through the list.
 
Thanks. Amy at Team Gleason sent me the following specs, so I'll try those first and see how I do, then fiddle as you suggested.
  • 44100 "project rate" (Hz)
  • export to desktop as a .WAV file and "signed 16 bit PCM" encoding (these settings come up during the Export process)
You can either record single messages, one at a time, or if you record a long string of messages, you will then have to edit them into individual ones.

(Nice forum software!)

I'm going to ask her if it matters what the files are called.

My husband has been a little confused about how to help. I'm going to enlist his help with the Audacity specs. I think that would be a good way to channel his energies and talents. He's been spending many hours reading cutting edge research about biomarkers, proteins that go haywire in ALS, etc. He hasn't understood that that is not actually the most helpful role I can imagine for him!

Can I let the audio files auto-name themselves, 001, 002, etc., or should they be named according to content? I will also ask Amy this question.

Boston suggests recording at myMessageBanking. Anyone have experience with that?
 
Status
Not open for further replies.
Back
Top