How to read frequencies from an audio file?

renegade

renegade

Staff member
Super Mod
Is there a way/program/snippet which can take an audio file as an input and read the frequencies? Basically I want to identify if a particular frequency exists in the audio file or not.

Is there an API available in J2ME platform which can help with this?

Sorry if I sound vague but I myself don't know much about this. Want to start exploring the possibility.
 
Not sure what you need it for.

I am also not sure why this in programming, so I may be a bit out of my depth here.

You can use a spectral analyser to check for presence/absence of specific frequencies. There are free spectrum analysers that will do it for you.

Here is one, but this needs a VST host (even Audacity or Foobar with a VST wrapper will work): Blue Cat's FreqAnalyst - Real Time Spectrum Analysis Plug-in (VST, Audio Unit, RTAS, DX) (Freeware)

If you need more granular approach (given that most naturally recorded sound will have almost all frequencies of interest present in some quantity) you will have to view the file yourself in an audio editor program such as Audition and look at the spectrum more closely.

Frankly, what you need is more of judgement call than software, as far as I can tell. A program will not be able to intelligently identify absence/presence. There is an audibility threshold for each frequency range.

Again, I may be out of my depth given this is a programming request. If you can explain the application I may be able to offer more precise suggestions.
 
I may be totally out of context here, but are you trying for audio fingerprinting? I mean I am trying to figure out the use case here.
 
The sound file will have a series of specific frequencies. Think of this as a morse code where the letters are frequency coded. So based on the series of frequencies the software should be able to output a word. Hence I posted it in programming.
 
Ah. So like tone dialling except with 26 or 26+9 characters, should be simple enough?

Single tones do not even need FFT (AFAIK), all they need is a 'if-then' algorithm to convert into a word if your tables for word to frequency are fixed.
 
Back
Top