Speed up speech

[louaychaer]

hi all

i want to acquire sound, then speed it up. i managed to speed it up but it sounds like a chipmunk. is there a way so that it wont sound like a shipmunk but rather maintain its original pitch.

Moreover i have to also make a VI that acquires a few seconds of sound then reverses the words. I can imagine that i need a way to define a delimeter so the software knows how to seperate each words but i cant manage to find how

and finally i want to make voice activate a switch: basically make a switch go on once a designated sentence is inputed.

any help is much apreciated

[Tamas_Szekely]

Hi louay chaer,

I found something (probably) interesting for you on the developer zone: Pitch shifting in Real-Time. http://zone.ni.com/devzone/cda/epd/p/id/6219

There are two more implementation: http://decibel.ni.com/content/docs/DOC-12605 and http://decibel.ni.com/content/docs/DOC-5181

I think after you speed up the waveform, you can shift down the pitch of the chipmunk speech to a normal voice 🙂

The examples shows which transformations are recommended.

I hope, I could help you to get closer to the solution.

Regards,

Tamas

[PaulieQ]

It's normal to sound like "a chipmunk" - this was good.

If you speed it up you increase the frequency of the sound/signal so it sound "high". If you slow it down it, the frequency is low so it sounds like a ghost in a scary movie.You have to do more than changing frequency.

Links provided by Tamas are interesting.

[louaychaer]

thank u for ur help guys

any other advise for the other problems im having??

[Tamas_Szekely]

Hi louay chaer!

I think, if you want to control a switch with voice, you have to do at least some pattern recognition in the sound signal, or do voice recognition.

I found you some links, maybe they can help you to find the best solution:

Some fundamentals: Using LabView for Voice Signal Analysis: http://zone.ni.com/devzone/cda/epd/p/id/5977

you can try also this, which uses external DLL to do the recognition: Voice Recognition in LabView http://sine.ni.com/devzone/cda/epd/p/id/4401ű

I heard that the Microsoft also released some DLLs, which can be also used to do it. (MSAgent)

Voice Recognition Data Acquisiton: http://decibel.ni.com/content/docs/DOC-4477

Voice Recognition using Advanced Signal Processing Toolkit:  http://decibel.ni.com/content/docs/DOC-2302

I hope I was helpful for you. I can recommend you to use the http://zone.ni.com search, because you can find a lot of interesting links about this topic 😉

Best Regards,

Tamas

[louaychaer]

thanks a lot for ur help