Ok, that sounds interesting. I don't know if the effort to edit the file will pay off, but I think it's worth at least looking at the file with a hex editor to see if there's enough information to fix any mistakes in the header.
About the file sizes, on page 137 of the H4n manual (the section about specifications), it says that the maximum file size is 2 gb. If you let the unit record for a long time, once it reaches the 2 gb limit in the file, it automatically spills over into a second file. So it looks like your first set was long enough to spill over into a second file, plus your second set was long enough, too. That's why STE-000 is exactly the same size as STE-002, both files were "filled up" to the limit that the H4n can record, then the recording spilled over into another file, for both the first and seconds sets. I don't know why STE-001 wasn't closed properly, because that should have happened automatically when you pressed Stop at the end of the first set.
Just for grins, let's take a look at the first file of the second set, STE-002. This should have lots of data, and should be nearly complete, because the recording had already spilled over into STE-003 by the time power went off. My general plan is to use a hex editor to examine the header of the .wav file. If it looks promising, I'd like you to use the hex editor to copy some of the header data to a response message that you can post in the forum (don't worry, it won't be too much data). Then I can examine the header data more closely to see what it looks like. If it looks like the header can be repaired, I will also later explain how to do it, although for now I'm going to leave out those details.
Go ahead and download the free HxD hex editor from here: http://download.cnet.com/HxD-Hex-Editor ... ?tag=mncol
. It's a zipped file, so you have to double-click the file to open the zip folder, then double-click setup.exe to install it. Then run the program, click File > Open, navigate to the proper folder on your PC, and double-click STE-002.wav to open it.
What you see next is a whole screenful of hexadecimal numbers, starting with the beginning of the file. Each row of numbers consists of 16 bytes of file data, and each byte consists of two hexadecimal digits, followed by a space. I don't know how familiar you are with hexadecimal, but it's just another way of counting things. Instead of counting 0 to 9, you count 0 to 15. The number system uses letters in addition to numbers, counting in this sequence: 0,1,2,3,4,5,6,7,8,9,a,b,c,d,e,f. The letters can be upper case or lower case, it doesn't matter.
On the left side of the screen, you see the "offset" value, which is the location in the file where that data row starts. The very first byte is at offset 00000000 and it counts up by 1 (in hexadecimal) for each byte after that, until the end of the file. The file might not have a multiple of 16 bytes, so the last data row in the file might not be full. On the right side, the program displays the text equivalent of each data byte. It's pretty common for files to have text strings embedded in them (including in .wav files), so this is a handy way of seeing the text without having to look up what each hexadecimal byte means. On the other hand, when you're looking at the audio data in the file, the text equivalent is not meaningful so it just looks like gibberish.
To navigate through the file, you can press Ctrl+Home on the keyboard to go to the first byte of the file, or Ctrl+End to go to the last byte. You can also use the Up, Down, Left, Right arrow keys to move around. Note that you can see a blinking cursor in the rows of hex data, plus you see a small box around the corresponding text character at the right side. Also note that when you use the left and right arrows to move around, the cursor moves by whole bytes. Whenever you modify data, it is modified one byte at a time, which requires you to type in two hex digits for each byte.
You can select data just like in a word processor, using the keyboard or using the mouse. Using the keyboard, you hold down Shift then move the cursor with the arrow keys. Bytes get highlighted to indicate they are selected. Using the mouse, you hold down the left button and drag for a similar result.
Once you have information selected, you can cut/copy/paste similar to a word processor, but in the hex editor we will not be using cut or paste, just copy. The usual "copy" command of Ctrl+C would copy selected information to the Widows clipboard, but in the hex editor, this just copies the selected bytes of data and not the addresses or text equivalent. This may be useful but not for us. Instead we will use an additional very useful "copy" command in the top menu: Edit > Copy as > Editor view. This copies not only the selected bytes, but also automatically includes the addresses and text equivalent.
Now let's take a look at the header; press Ctrl+Home to go to the beginning of the file. A .wav file is an old file format from way back when, and it uses a general file format known as RIFF. The general structure of a RIFF file is to have one or more "chunks" of data. Each chunk has a identifier at the beginning saying what type of chunk, followed by some other data that varies depending on the chunk type.
The first four bytes in the file should be the letters "RIFF", which you should see in the text area. Then there are four bytes of gibberish, followed by the text letters "WAVE", which lets you know it's a .wav audio file (there can be other types of RIFF files). Following that, my H4n puts a chunk that is a Broadcast Wave Format extension. This starts with the letters "bext", followed by a bunch of data for that chunk including some text strings. If you scroll down through the file, down around address 360 or so (it might be different on your file), you should see the letters "fmt " (note the trailing space) which is a chunk of data that identifies the particular type of .wav file (the various audio parameters). A little later in the file you should see the letters "data", which is the chunk that contains the actual .wav audio data. On my H4n audio files, the .wav data continues right up to the end of the file, there are no more chunks after it.
So I'm hoping that you see some or all of these chunks in your file, which would be a good sign. If you do, then I'd like you to copy some information to a response message that you post in the forum.
First, using Windows Explorer (the file browser), right click STE-002.wav, then click Properties. In the General tab, copy the exact byte count from the Size field (you can just drag the mouse cursor to highlight, then copy/paste into the response message). Then in the Summary tab, copy down all the audio parameters. If these are not correct, please indicate what sample rate and sample size you were using (for example, 44.1 kHz at 16 bits, stereo).
Next, in the hex editor, scroll the cursor down until it is a few rows below the beginning of the "data" chunk. Make sure the cursor is at the start of a row (press Home), then hold down Shift and scroll up to the beginning of the file, which selects the .wav header. Now use the special copy command, Edit > Copy as > Editor view. Paste this data into your response message.
Finally, scroll to the end of the file (Ctrl+End), hold down Shift, then move up a few rows to select a few rows from the end of the file. Use the special copy command then paste the data into your response message.
Well, I think that's enough for now. Once you post the response with all the data, I'll take a look at it. I hope the file is in good enough shape. It will take me a little while to look at it, and I'll get back to you with more information. By the way, I am on the East Coast of the U.S. and we are having another major winter storm, so if my power goes out, it could take a few days.
- John Santic