Oh no OMG

 

Wow so you want a bigger file with the same poor audio quality. Why?

 

No, he doesn't, as he explained here

 

Thanks God.

 

So we're on page three answering a question to which the answer was simply "no"

 

Yes!

 

So I know this post is 9 years old but I am going to be a pedant.

Lossy format - When the file is compressed (made smaller) it is done by physically removing / altering the data the file contains.

Lossless format - When the file is compressed, it is done in a series of steps that are reversable.

Therefore:
Lossy (in which unnecessary information is discarded) - loses information from the original file to make it smaller, the original can never be recovered.
Lossless (without loss of information) - there is no physical removal / permanent alteration of the original file to make it smaller so the original can be returned to by following the reverse of the compression process.

Think of it like a sponge. I can compress the sponge in my fist and make it smaller but when I release it, it will return to the original sponge. If I bend the sponge in half, very carefully cut around the middle and hollow out each end, now I can compress the sponge into a much smaller space but when I release the sponge - however much it may look like it - what I end up with is not the original sponge and I can never go back to it.

Take the binary string - 00001111001100000111001111100111110000111110000

Lossy - just half every group of binary numbers and round down - 001101001011011001100
If I reverse this by doubling every group - 000011110011000011001111000011110000, not the same as the original

Lossless - count each bit occurence - 40412021503120512051405140 (in reality this would be incrediby wasteful as each consecutive number of bits would have to be counted using a 3 bit Binary number 000 = 0, 001 =1, 010 = 2, 011 = 3, 100 = 4, 101 = 5, 110 = 6, 111 = 7 so the actual data would be (100)0(100)1(010)0(010)1(101)0(011)1(010)0(101)1(010)0(101)1(100)0(101)1(100)0 or bigger than the original but this is merely an example).
If I reverse this process I get 00001111001100000111001111100111110000111110000 or the original
4 4 2 2 5 3 2 5 2 5 4 5 4

Hope that helps.