Perhaps I should rev it up to the highest quality setting? WARNING: This song would at least be 120MB if not more if I did that.
Regardless, I doubt the highest quality setting Goldwave can create would make a 120mb audio file which is only three minutes long...an absolute killer sample rate maybe but that's definitely overkill.
I totally agree with what Petrie said. There should be No need for a sampling rate higher than 44.1khz (44100 Hz). I can't guarantee that I can physically add sampling rates higher than that into our song anyway. If a .wav file is over 30 MB, do something with it. There's something wrong.
:
I can't think of any other reason that the quality could possibly be increased, other than increasing the sampling rate, so I wouldn't worry about increasing the quality.
From my perspective, the solution is simple: Get a better microphone, if you have the money. Though I must admit, I don't understand the problem with your karaoke machine's microphone, it is doing strange things to your recordings, especially with the buzzing.
Your other microphones simply will be incapable of recording at good quality. Built-in laptop mics are basically for chatting over instant messengers (VOIP) and recording lectures; that's basically it. They're built for power conservation and convenience, not quality.
I am currently using the program Action9000 recommends.
My little tutorial explains the basics on what I'd recommend doing in Goldwave. No other actions should be necessary. Maybe, however, I can make Goldwave a bit clearer for you.
I assume that you followed my instructions on making a new file exactly as written:
Make every window related to Goldwave full-screen. When the new file opens up, you will see a window containing a long green line. This line represents the motion that will be sent to your speakers when the file is played. A horizontal line indicates no movement. This line is a visual represenation of the sound waves that you will ideally hear (from a perfect set of speakers, which obviously don't exist). This line is known as the
"waveform".You will notice two scales in goldwave; a horizontal scale (time) and a vertical scale (amplitude/volume). The vertical scale (Y-Axis) has a range from 0.0 to 1.0 in both the positive and negative directions. This represents the total RANGE of volumes that the digital audio file can capture and store. 0.0 represents silence and +/0 1.0 represents the maximum amplitude from the 0.0 point. Therefore the loudest sound you can possibly record is a wave that travels from 0.0 to +1.0 to -1.0, back to +1.0, back to -1.0, etc.
The level that Goldwave percieves when recording from a microphone is related to the microphone volume level ("Gain"). When a sound is recorded, the vibration of air around the microphone is converted to a digital signal and sent to the computer, which stores the amplitude as a 16-bit value (in 16-bit audio). The value of this 16-bit number is dependant on a mathematical calculation containing your input volume (the volume of your voice and surrounding noise) and the gain of the microphone. I'm assuming you understand the basics of binary, being a computer repair guy, so I'm going to point this out:
The minimum binary value in a 16-bit number (representing Silence) is obviously
0000 0000 0000 0000
The maximum binary value (representing maximum possible volume) is therefore
1111 1111 1111 1111
Any binary number between these is some amplitude between 0.0 and 1.0. We are using SIGNED 16-bit audio so the final bit is technically the sign of the number (positive or negative), which is why we see +/- 1.0 on the Goldwave scale.
When setting the microphone gain, we ideally want to get as much of your voice as possible and as little background/electromagnetic (EMR) noise as possible. This ratio, Signal to Noise ratio, is the biggest problem with your recordings. Ideally the sound would be 100% signal and 0% noise but this is impossible. Our goal is to get as close to this as possible. This is done by using high-quality equipment and setting the microphone gain to an appropriate level.
What is the appropriate level? Ideally, you want the LOUDEST instant of your voice to be as close to +/- 1.0 on the Y-Axis scale without going Over 1.0. If any instant of sound is calculated (based on microphone gain and your voice volume) to be LARGER than +1.0 or SMALLER than -1.0, the 16-bit number cannot store it (where would it go? We are out of digits! You can't put a larger value than 1111 1111 1111 1111 into 16 binary digits.). As a result, it must be assigned to the most logical place: the nearest possible binary value. As a result, a sound that continually calculates to larger than +1.0 or smaller than -1.0 will result in the Cutting off of the peaks of the wave form (because the peaks get flattened to the 1.0 value). This flattening distorts the sound and results in a rather unpleasant noise.
This, however, is
not a problem I have been hearing in your recordings, LBTlover. Instead, I am hearing unusually loud levels of hissing and background noise.
Here is the problem I am seeing with your recordings:
By the way, the 0.04 is a Maximum noise level, so that is basically the lowest SNR in your entire recording. Most of the time it may be closer to 10 or even 15 dB. That is still extremely low.
As for what to do about it:
I would say Increase the gain on your microphone. Your Maximum volume right now at Any point is about 35% or 36% of the 16-bit maximum. You can crank that gain quite a bit higher. If that doesn't help, you may need either a better microphone or a better sound card.
If you can get that Signal to Noise ratio better, I can do the rest. B)
