I suspect that, though I haven’t used it much.
Plug-in developers are welcome to make new plug-ins for Audacity, and it is also possible to create add-on “modules” for Audacity, though I believe this is still in its early days.
So probably no plugins would be needed - just interface additions.
like in the following program?
Transcribe!
I doubt that will ever be supported as it is a standalone commercial program, so unless the author is prepared to release the source code under a GPL compatible open source license, or create a plug-in for Audacity, then it cannot be used in Audacity. Their software licence does not allow anyone else to modify the program to allow it to work with Audacity, so it’s stand-alone or not at all. (Long live Open Source > 
> )
I just mean the features, mainly the interface, as Audacity may already have the underlying functionality.
Also, there are some musically oriented actions, such marking up the piece into sections, measures, and beats,
I would LOVE to see that feature. It would open up the power of Audacity to a whole different type of music production.
There are already these features on the > “Feature Request” page of the wiki
_* BPM and beat timecode detection with automatic beat matching (35 votes)
- Then let the Timeline display the detected time signature and bars (4 votes) >_
Would you like me to add your votes?
Yes, this is what I mean. The program I mentioned above has a nice interface. The interface itself is another direction of Audacity as mainly an editor. But such an interface for analyzing may also be helpful when editing.
Automatic beat detection can at most be a tool for a preliminary markup in a music context. In Balkan music, a group of musicians may not agree on the exact location of the beat when getting down to time bends measurements - somewhere between a 1/10th and a 1/100th of a second, or just slightly above what is audible. So one has to focus on perhaps one musical instrument, which may even be different from beat to beat. The beats happen unevenly, and may not be clearly marked by the musicians. and in a musical piece, there are several structures one would want to mark: sections, subsections, measures, beats, sub-beats.
Also, there is no unique way to display a time signature. One may write a piece in 2/4 time with triplets. Beethoven’s 5th is normally played in one though written in 2/4. Due to time bends, it is hard to assign a meter to the Bajrace - 12 = 3+2+2+3+2 with quadruplets on the 3s is one possible suggestion.
So such features can easily get in a way if they cannot be disabled, though it might be fun to see what it suggests.
The above mentioned program has a feature where it suggests chords. In Balkan music, it mostly displays no chord, but says “out of tune notes”. But the feature might be nice, if it could display which tones it can pick up. And possibly change the tuning system it uses as reference.
There is a primitive keyboard, for playing pitches.
Audacity is not a real-time music application, so without a total rewrite you will not get real-time keyboard input. (A total rewrite essentially makes it a different program and no longer “Audacity”).
However, there are feature requests for (non-real-time) midi support and it has its own section on the Feature Request wiki page > Missing features - Audacity Support
I did not have MIDI support in mind, here, as one can use other programs for that, like Scala Scala Home Page. Also, I wrote a program for ChucK http://chuck.cs.princeton.edu/ by which one can play different tunings systems (Pythagorean, E12, meantones, etc.) using the typing keyboard. So if implemented at all, this would be something very basic.
It is sometimes useful not having to work with another program, though. The one in Transcribe is oriented at displaying the pitches in E12 and cents. So the use would be that one wants to know what the played pitch is by listening to a comparison - graph might not be a good indicator in real life music.
Also, there is this program, which not only presents the spectrum, but one can change it, and listen to it
That looks interesting, but no, it’s not open source. If you have tried this program you will have agreed to the license conditions, which you must do > twice > when you install it.
The situation might be more hopeful - the copyright owner just hasn’t made the sources public. I think I saw somewhere that it is a revival of some older program.
It’s not exactly showing the “spectrum”, it is analysing “partials” in the music and allowing you to manipulate them in various ways. This program could theoretically be used to accomplish the illusive “vocal isolation” that everyone wants (though it would require a great deal of manual editing to do so).
Horizontally there is the time line, and vertically, the frequencies. Then this graph can also be manipulated, and one can listen to the changes. I haven’t used it much, but the feature looked interesting. It would be more useful if the partials could be analyzed more in detail, relative a tuning system - I think it just displays raw frequencies in Hz, which is not so useful in a musical context. Just mentioning it.