06-record-table.py - Recording live sound in RAM.
============================================================================================================================================


By recording a stream of sound in RAM, one can quickly re-use the
samples in the current process. A combination NewTable - TableRec
is all what one need to record any stream in a table.

The NewTable object has a `feedback` argument, allowing overdub.

The TableRec object starts a new recording (records until the table
is full) every time its method `play()` is called.

.. code-block:: python

    from pyo import *
    import os
    
    # Audio inputs must be available.
    s = Server(duplex=1).boot()
    
    # Path of the recorded sound file.
    path = os.path.join(os.path.expanduser("~"), "Desktop", "synth.wav")
    
    # Creates a two seconds stereo empty table. The "feedback" argument
    # is the amount of old data to mix with a new recording (overdub).
    t = NewTable(length=2, chnls=2, feedback=0.5)
    
    # Retrieves the stereo input
    inp = Input([0, 1])
    
    # Table recorder. Call rec.play() to start a recording, it stops
    # when the table is full. Call it multiple times to overdub.
    rec = TableRec(inp, table=t, fadetime=0.05).play()
    
    # Reads the content of the table in loop.
    osc = Osc(table=t, freq=t.getRate(), mul=0.5).out()
    
    
    def saveToDisk():
        savefileFromTable(table=t, path=path, fileformat=0, sampletype=3)
    
    
    # After two seconds, the table content is saved to a file on disk.
    sv = CallAfter(saveToDisk, time=2).play()
    
    s.gui(locals())

