To be clear, converting a .wma file to a text file is technically impossible. The two reside in completely different worlds, like apples and oranges. However, there is a way to turn spoken words into text using voice recognition software.
A .wma file is a Windows Media Audio file. This means it is audio that is encoded to be played specifically by Windows Media Player. A text file is a file you would open with text editing software, such as Microsoft Word. Windows Media Player cannot open a text file, just as Microsoft Word cannot open an audio file. The two file types are different, thus they cannot be "converted" from one to the other, or vice-versa.
The only way to convert the audible information found in a .wma file into text would be to use voice recognition software. This type of software uses technology that has been researched since 1936, when it was referred to as "speech recognition." The software itself is programmed to recognize speech patterns and syllabic sequences and compare them against a vast library of stored information. The software determines what is being said, and then it transcribes the spoken words to text.
Now that you understand how voice recognition software works, you can probably guess that it would not work very well with music. The amount of extra noise behind any lyrics would confuse the software, and it would not function properly. If you have a .wma file, and you want to transcribe it to text, it would need to be a simple, clean recording of the spoken word.
Download and install voice recognition software. A list of reviewed software choices can be found here: http://www.consumersearch.com/voice-recognition-software
Open the .wma file and make sure it is clearly playing through your computer speakers.
Place a microphone in front of the speakers so that it is capturing the audio coming out, and make sure your voice recognition software is detecting the audio.
You will know very quickly whether or not the voice recognition software is functioning well enough for your purposes.
Things You'll Need
Voice Recognition Software