I do shows with recorded music, and i depend on internet sources for material. My question is whether recording audio from videos with Audacity may be better than downloading and importing the audio. I recently learned that the highest quality you can download from most Youtubes is lower than the highest you can stream, if you set it to highest. If (I know it’s a big ‘‘if’’) the audio of the stream is better than the downloaded version, am I better off recording the stream? How much loss results from recording? The reports I’ve read, about how the audio does or doesn’t follow the level of video HD, are confusing. I wonder if there is any way to measure this.
In Maine? New Hampshire?
I don’t know of any way to say that this clip “was encoded at quality 64.” Particularly since, as in your case, the sound may have been compressed multiple times. Any test we might do tell may only tell you the last encoding, if it worked at all.
The only semi-definitive way to determine the quality is to wreck it.
If you encode a perfect quality mono show (that you shot to WAV format in your garage, for example) at 32 (MP3), the show will have some very minor artifacts, but most people will be able to listen to it all day with no trouble. If the show turns to garbage immediately, then it wasn’t perfect quality to begin with. It was probably already compressed. How much and how many times will depend on how high you have to go in your test encoding to make the show sound OK.
If you make it all the way up into 256 and 320 and the show still sounds a little funny, then it probably was 32 all along, or was taken from a show that was 32. 32 is a magic place. The slightest bit lower quality and most people can hear distortion and damage. It’s 64 for stereo.
So prepare your two clips and see which one dies first (save the originals, obviously).
Numbers like 32 and 64 are pretty depressing for me. I want to exceed 128 anyway. I read that “HD” videos are usually about 160 kbps now. I understand that there is a lot of compressed trash on Youtube, but there are also what appear to be professionally produced videos. I have learned to recognize these, though not perfectly I’m sure. Many of them are downloadable at 720p by 1280p maximum video rating, but streamable at 1080p by 1920p. I am having a hard time finding out whether the audio is correspondingly rated. But assuming a stream is higher kbps than the download, my question is whether there is any advantage to recording with Audacity rather than downloading and importing to Audacity. Is there enough loss in the recording process to reduce the difference?
Maybe I need to isolate and clarify the question.
How much loss is there when recording with Audacity? How much higher kbps audio stream, relative to kbps of a download, is necessary for there be an advantage to recording the higher stream?
Recording with audacity doesn’t cause any damage to the signal. Neither does downloading the video and extracting the audio. I prefer the second method cause it’s easier and faster.
What might induce some damage, is using Windows.
And what ruins the compressed Youtube sound, is uploading any audio that peaks over -3 dB. A second thing that is wise, if you really need to compress your audio before uploading to YT, is using a multiband compressor. You really need different settings for bass and high (transients).
What completely ruins sound, is using mp3 as upload. Upload uncompressed wav if you can.Youtube compression in itself isn’t as bad as many people believe.
Aren’t you producing a music production for a North-East radio station? It’s not hard to bump up against the 64 quality wall. As I recall, you’re downloading music and creating a show at home and then the station required either MP3 submission, or Audio CD. We strongly recommended the CD. The station broadcast it OK, but then posted your show on their web site…as an MP3.
And you lose on SportsCall®. All those conversions cumulatively slid under the stereo 64 boundary and the complaint was bubbly, honky sound.
Are you bicycling the CD to the station? That eliminates one conversion right there. Bicycling is the metaphor. I don’t think anybody is doing a lot of bicycling in Maine right now.
As long as you capture the work in Audacity the internal quality is almost certainly to be far higher than anything you download. I still think the only way to definitively tell which version is a better download is to wreck them. Encode them tiny (as a test) and don’t use whichever turns to trash first. That is really the only test. All of these sound fine as long as you don’t line up the compression steps like that.