• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

How to convert the human voice into digital form

Status
Not open for further replies.

plotoyee

New Member
i'm now doing a sound remote control. but im facing problem in converting the voice in to digital form. can any one tell me which IC that i can used to solved this problem.

if any one have done this project before.
can u please send to me the information

thanks a lot

:D
 

kinjalgp

Active Member
Yes, you have to use medium/high speed Analog-to-Digital converter having sampling frequency of atleast 40kHz for accurate sound reproduction of sonic signals (20Hz to 20KHz). Thats what you need.

For near real-time processing of audio signals use PIC or any other high speed microcontroller of your choice. For real-time processing go for Digital Signal Processor (DSP). You'll find lot of information on DSP's at www.ti.com .

This was all about conversion of audio into digital form with good quality.

Since you are doing a remote control project, you don't need very high quality sound for detection. I had found one such project. Let me search my disk, if I can find the link. It is based on motoral 68HCxx uC.
 

kinjalgp

Active Member
I found the document but not its link. It was downloaded from www.circuitcellar.com . The project name is "Low Cost Voice Recognition". Try searching the site.
 

Manan

New Member
analog to digital converter

I have a circuit for a analog to digital converter... I just completed the project for my design and manufacturing class so if u want u can use it …

The chip (IC) I used was a dual 4 bit binary counter (74LS393N) made by Texas Instruments. And one (741) Op-amp. E - Mail me for the schematics
 

crust

Member
kinjalgp said:
Yes, you have to use medium/high speed Analog-to-Digital converter having sampling frequency of atleast 40kHz for accurate sound reproduction of sonic signals (20Hz to 20KHz)
For music applications 20-20k is considered the desired range. However, that is way beyond what is needed for an audio remote control -- the sound will not be played back. The spectral content of human voice at higher frequencies is very low. The bandwidth of your average telephone is only 3kHz. The important part of this project is to recognize the voice commands, not reproduce them.

I think plotoyee could get by without sampling at 40kHz, I would start at telephone sampling (6kHz) and go down from there. I think the actual algorithm is probably going to be more difficult than determining the proper sampling frequency.
 

kumar_3k

New Member
useful link to search about audio recognization...

hey man u can use ADC chip...such as ADC0804...for futher information u can view ADC data sheet ...lot kind of ADC available by National Semiconductor , even for audio recognization. http://www.national.com please serch here.........thanks.
 

kinjalgp

Active Member
crust said:
kinjalgp said:
Yes, you have to use medium/high speed Analog-to-Digital converter having sampling frequency of atleast 40kHz for accurate sound reproduction of sonic signals (20Hz to 20KHz)
For music applications 20-20k is considered the desired range. However, that is way beyond what is needed for an audio remote control -- the sound will not be played back. The spectral content of human voice at higher frequencies is very low. The bandwidth of your average telephone is only 3kHz. The important part of this project is to recognize the voice commands, not reproduce them.

I think plotoyee could get by without sampling at 40kHz, I would start at telephone sampling (6kHz) and go down from there. I think the actual algorithm is probably going to be more difficult than determining the proper sampling frequency.
I think you haven't read my post completely. I've already mentioned that you don't need high sampling rate for detection. :)
 

plotoyee

New Member
project that i'm doing

i'm now doing a project about "control by using human voice"
do anyone done this before
 
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top