Hi there,
Thanks for the updates to the program (2.2 installer)!
According to the Wikipedia page the file sizes vary according to the bitrate. However it seems to be stuck on 44100.
It’s possible to resample at a lower rate but the file size of the export wav remains the same.
And no matter what the sample rate of the raw file, the resample box in menu still defaults to 44100.
But then it’s working with ye olden 8 bit clips. Problem there? Is there a chart or something that explains bit rate / versus bit resolution?
Aha: Oops it’s the Project bit rate set at the bottom of the screen that governs. Sorry missed that. working now!
According to the Wikipedia page the file sizes vary according to the bitrate.
That’s true. The bit rate is different from the sample rate. The bit rate is always directly related to file size. With compressed files the sample rate is not directly related to size.
The bit rate is usually in kbps, which is kilo_bits_ per second. Normally, we use the bit rate when dealing with compressed files, although you can calculate the bit rate for uncompressed files. If you know there are 8-bits in a byte, you can (approximately) calculate the file size from the bit rate and the playing time. There is some space used for the file header(s) and any metadata (artist/album/title/artwork “tags”) can add to the file size. Album artwork can add significantly to the size of an MP3.
The sample rate is the number of samples per second. You can calculate the file size of an uncompressed file by multiplying Sample Rate x (Bit Depth/8) x Number of Channels x Playing time in seconds. Or, you can calculate the bit rate - The bit rate for a CD is 44.1kHz x 16 bits x 2 channels = 1411.2 kbps.
And no matter what the sample rate of the raw file, the resample box in menu still defaults to 44100.
Hmmm… When I open an MP3 ripped from a DVD, it correctly shows 48,000.
Problem there? Is there a chart or something that explains bit rate / versus bit resolution?
Sample rate and bit depth are independent. …I can make a 44.1kHz file at 8-bits, 16-bits, 24-bits, or 32-bit floating-point, etc.
Generically, the sample rate for television is 48000. The one for Audio CD is 44100. The TV one came second, they had the experience and they had a lot more room to mess with, and so could use higher numbers. 44100 is the lowest sample rate that you can shoehorn into a good quality music system and still have most people not be able to tell what you just did.
Remember when they were doing the flat, shiny disk, MP3 didn’t exist yet. Your choices were top, perfect quality or go home. There was no middle ground like there is now.
The object is to whack up the sound into digital chunks, do something with it and then listen to the show at the end. 44100 is the lowest sample rate which will do that, and even then takes a trick or two for it to hide its petticoats. If you do the strict arithmetic, 44100 is only surgically accurate to 17,000 audio, not the generally accepted top end of 20,000. That’s one reason people in higher end sound production drop it as a hot rock in favor of higher sample rates such as 96000.
Once compressed sound flexed its muscles, then you could divorce bitrate (fixed for an older audio CD) from sample rate.
The worst problem with that is the general assumption all sound is MP3 and all MP3s are the same…and free.
Hmmm… When I open an MP3 ripped from a DVD, it correctly shows 48,000.
Yep, the clip that was the problem was the extracted MP3 from Fantasy Island Intro
Even though it’s sampled at 44.1 the resolution is very poor and mono. Reducing it to 22050 or even 11025 made no difference to my ears. As did the conversion to 8 bit sound. The original MP3 was 1.31mb so converted to mono, amplify to defaults, changed project rate setting to 22050 (important), resampled at 22050 and saved as 8 bit aiff with MS wav header.
1.80 mb. Pack that in 7z with max compress and the file is only .98 mb, actually a lot more that supposed.
But reducing to 11025 it’s only 926kb. Compress to 7z is better though: 616kb.
Interesting. Is there more info on that somewhere? What’s 96000 accurate to?
This is the official download page on the Audacity website: http://www.audacity.audio/download/
We cannot guarantee the authenticity of downloads from other websites, so we strongly recommend using the official.
2.1.2.0 is not a genuine version number either. The actual version number can be found by selecting “About Audacity” from the Audacity “Help” menu.
I’d guess that you are using “Audacity 2.1.2” but if it says “2.1.2.0” in “About Audacity”, then the version that you are using has been tampered with.
I wish you would stop writing that koz - it’s NOT TRUE.
With modern anti-aliasing filters, 44100 Hz sample rate PCM audio is “surgically accurate” to about 20,000 Hz.
Way back when CDs were first invented, anti-aliasing filters were not so good, so back then your statement was more realistic, but “modern” digital filters that are very close to “ideal filters” have been used for well over a decade.
“Theoretically”, digital audio (PCM) is capable of accurately reproducing frequencies up to half the sample rate. So for 44100 Hz sample rate, the theoretical frequency limit is 22050 Hz. This theoretical limit is called the “Nyquist frequency” (https://en.wikipedia.org/wiki/Nyquist_frequency).
In practice the limit is a little lower because when digitizing an analog signal it is essential that frequencies above the Nyquist frequency are removed before the signal is sampled. The practical limit therefore depends on how good the filters are that remove frequencies above the Nyquist frequency. Modern digital filters are very good and run up to about 95% of the Nyquist frequency, so that gives you a practical upper limit close to 21 kHz for a sample rate of 44100 Hz.
For 96000 Hz sample rate, the practical upper frequency limit ‘could’ be up to about 45,000 Hz. However there are disadvantages to retaining ultra-high frequencies in audio recordings. For example, one of the disadvantages is that loudspeaker drivers either fail completely of are very inaccurate at such high frequencies, and this can cause “Intermodulation distortion” (https://en.wikipedia.org/wiki/Intermodulation), so including frequencies that are beyond the audible can reduce the sound quality within the audible range. For this reason the frequency response of audio equipment may be deliberately limited to a range much lower than is technically possible. Unless the manufacturer is wanting the hype of big numbers, it is likely that frequency range will be limited substantially lower 45khz and could be little over 20 kHz (which is still sufficient for “audio”).
My understanding is, having also (http://www.codeproject.com/Tips/1089681/Bigger-Directories) published software, that right click Properties/Details of the executable file is where the correct versions are displayed. And Cnet is (and has been for years) a reputable source for files, and AFAIK they do not publish material unless it has the explicit permission of the author.
Buuut, as using IE11 on Win 8.1, I did first try the official site. The actual file does not download, sorry.
To avoid confusion about version numbers there is a guideline given in the pink box at the top of the page:
Please state which version of Windows you are using, the exact three-section version of Audacity from Help > About Audacity
Regardless of the reputation of mirror sites, we are not able to guarantee the authenticity of downloads from any site other than from our official download partner.
I’ve just checked and it downloaded perfectly for me. If you can describe precisely what you clicked on, then I can check it out.
Cnet isn’t a reliable source at all. As most of these downoad sites, they tried making money from other people’s work, by wrapping their programs in an adware tool. See what the EFF has to say about it:
That is correct. As long as you don’t see it in Help > About Audacity…, 2.1.2.0 is the correct full version number which you can see in the audacity.exe properties or by hovering the mouse over audacity.exe.
The “0” comes from #define AUDACITY_MODLEVEL 0 in src/Audacity.h.
