Notices
Results 1 to 8 of 8

Thread: WAV Encoding Samples Interpretation

  1. #1 WAV Encoding Samples Interpretation 
    New Member
    Join Date
    Jan 2012
    Posts
    4
    Something I've never quite figured out- what do the data values in a WAV file's samples correspond to?

    I understand that there are chunks (I've already used a hex editor to look through the FMT chunk to see that its uncompressed/1-channel/22kHz sampling rate), and that the DATA chunk simply consists of sample values, but I don't know what those sample values mean.

    For example, if you create a vector of sample values in MATLAB using the WAVREAD() command, all the values are between -1 and 1; are these ratios between maximum and minimum voltage of the mic's analog input? Or ratios of some reference level? If you feed this vector into SPECGRAM() function, you get a nice FFT'd spectrogram but I don't know what hand-waving is being done here (because I don't know what the original values represented).

    Is there a simple answer here? Or is my thinking way off base to begin with?


    Reply With Quote  
     

  2.  
     

  3. #2  
    Forum Freshman
    Join Date
    Nov 2009
    Posts
    52
    You start with the IETF, (probably starting with RFC 3625 and moving on from there) - Always the first stop in getting to grips with a standard protocol.
    ...
    At a glance: Wave File Specifications


    Reply With Quote  
     

  4. #3  
    New Member
    Join Date
    Jan 2012
    Posts
    4
    I think my lack of understanding is on a more basic level; I've read the spec (which is why I understand about the FMT and DATA chunks and how they're laid out) but don't understand what the discrete data values themselves represent.

    I asked someone locally who had done some Sound Engineering, and the description I was given was that the -1:1 range of the sample values indicated a ratio of mic input voltage, and that could then be converted up to "line voltage" (5V) to be comparable between mics. For anyone who has a clearer understanding, let me know if this is off base.
    Reply With Quote  
     

  5. #4  
    Comet Dust Collector Moderator
    Join Date
    Mar 2011
    Location
    New Jersey, USA
    Posts
    2,848
    The coding is for an instantaneous voltage value of a complex of sine waves that can be converted by a speaker (or headphone) to sound waves.

    No mic is necessary; for example, and electic guitar or organ, or many other devices can transmit the raw electrical signal that it uses to create sound without the actual creation of sound being necessary.
    Reply With Quote  
     

  6. #5  
    New Member
    Join Date
    Jan 2012
    Posts
    4
    Ok, that makes sense. From that viewpoint, how does the voltage input correspond to the output value? For example, does 0V correspond to 0.0 and 125 mV correspond to -0.378, etc (I don't think those numbers go together, I'm just trying to clarify my thinking)?
    Reply With Quote  
     

  7. #6  
    Comet Dust Collector Moderator
    Join Date
    Mar 2011
    Location
    New Jersey, USA
    Posts
    2,848
    Well, I haven't read the standard, but there's a maximum value and a minimum value. Everything after that depends on the devices that actually convert the signal to sound.
    Reply With Quote  
     

  8. #7  
    New Member
    Join Date
    Jan 2012
    Posts
    4
    When looking over the standard, the min and max I found were for the integer values (which is pretty obvious; for example, -32XXX to 32XXX for a signed 16 bit number). I didn't see anything referring to the measured values.
    Reply With Quote  
     

  9. #8  
    Comet Dust Collector Moderator
    Join Date
    Mar 2011
    Location
    New Jersey, USA
    Posts
    2,848
    They are not absolute values. They represent signal values that are intepreted by the output setup to recreate the sound.

    If a device needs 0-1 volt input to the sound setup, that's how it reads the digital values, if it needs 0 to 125 mv input, that's how it reads the values, if it needs +/ 5 volts, that how it reads the digital values. There's a converter in between that takes the digital values and creates the analog voltages needed to drive the amplifiers and speakers properly.

    Hope that helps.
    Reply With Quote  
     

Similar Threads

  1. interpretation
    By parag1973 in forum Mathematics
    Replies: 5
    Last Post: August 28th, 2011, 10:56 AM
  2. Born interpretation.
    By sox in forum Physics
    Replies: 2
    Last Post: August 27th, 2010, 09:09 AM
  3. Searching Samples for X-Ray Analysis
    By Karanum in forum Physics
    Replies: 4
    Last Post: April 28th, 2010, 06:03 PM
  4. Dream Interpretation
    By marcusclayman in forum Behavior and Psychology
    Replies: 24
    Last Post: August 11th, 2009, 05:45 AM
  5. dna data samples
    By tomd123 in forum Biology
    Replies: 3
    Last Post: March 28th, 2008, 04:40 PM
Tags for this Thread

View Tag Cloud

Bookmarks
Bookmarks
Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •