LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Raw PCM to WAV

I very much appreciate your response.

 

I actually did/had gone through to the best of my ability. I realize that the ChunkSize is identified via your import, but I can't seem to understand the rest of your header. Following WAVEfmt, you coded:

"10.00.00.00" for "16" for SubChunk Size
"01" for AudioFormat
"01" for Mono Channel

 

"80.3E.00.00" for "16000Hz" - which I expected "00.00.3E.80" or "3E.80.00.00" (based off my basic understanding of encoding)

 

You then have "00.7D.00.00" for Byte Rate

 

"04.00" for Bits per sample, which I don't understand? You are specifying 4 bits per sample?

 


Thanks greatly for any clarification you can offer...

0 Kudos
Message 11 of 15
(308 Views)

some parameter doesn't affect (in my case).

here corrected:

Positions Sample Value Description


0 - 3 (0x52,0x49,0x46,0x46)"RIFF" ==> Marks the file as a riff file. Characters are each 1 byte long.
4 - 7 (0x32,0x4F,0x00,0x00) = 0x00004F32 = 20,274 ==> File size (integer) Size of the overall file - 8 bytes, in bytes (32-bit integer). Typically, you'd fill this in after creation.
8 -11 (0x57,0x41,0x56,0x45 = "WAVE") ==> File Type Header. For our purposes, it always equals "WAVE".
12-15 (0x66,0x6D,0x74,0x20 = "fmt ") ==> Format chunk marker. Includes trailing null
16-19 (0x10,0x00,0x00,0x00 = 0x00000010 = 16) ==> Length of format data as listed above
20-21(0x01,0x00 = 0x0001 = 1) ==> Type of format (1 is PCM) - 2 byte integer
22-23 (0x01,0x00 = 0x0001 = 1)  ==> 1 Number of Channels - 1 byte integer
24-27 (0x80,0x3E,0x00,0x00 = 0x00003E80 = 16,000 ) ==> Sample Rate - 32 byte integer. Common values are 44100 (CD), 48000 (DAT). Sample Rate = Number of Samples per second, or Hertz.
28-31 (0x00,0x7D,0x00,0x00 = 0x00007D00 = 32,000) ==> (Sample Rate * BitsPerSample * Channels) / 8 ==> (16000 * 16*1) /8 = 32,000
32-33 (0x02,0x00) = (16*1)/8 = 2 ==> where (BitsPerSample * Channels) / 8. 1 = 8 bit mono, 2 = 8 bit stereo/16 bit mono4 - 16 bit stereo
34-35 (0x00,0x10 = 0x0010 = 16 ) ==> Bits per sample
36-39 (0x64,0x61,0x74,0x61 = "data")  ==>where "data" chunk header. Marks the beginning of the data section.
40-43 (0x00,0x00,0x4F,0x0E = 22238) ==> NO Little Endian. File size (data) Size of the data section.

 

see vi modified.

 

0 Kudos
Message 12 of 15
(300 Views)

This helps a lot!

 

On line 34-35, did you mean: "10.00" instead of "00.10" ?

And on line 40-43, what do you mean "NO Little Endian"?

 

I'm very grateful, thanks!

0 Kudos
Message 13 of 15
(296 Views)

sorry, you are right (is in Little Endian)

do you know what Little Endian means?

please take a look here: https://en.wikipedia.org/wiki/Endianness

 

and the next is in Little Endian too, my mistake

and here: http://soundfile.sapp.org/doc/WaveFormat/

 

 

 

 

but don't worry about that

 

on line 40~43, no need to put use Big Endian

 

0 Kudos
Message 14 of 15
(292 Views)

I understand the differences a little.. it's not my specialization.

 

 

Those lines HAVE TO BE little endian? I think my PCM is big endian... but I have no idea because my source is foreign. I've set everything else up the same way. I think the data is big endian too.

0 Kudos
Message 15 of 15
(290 Views)