I think it's because there is an infinite loop (while True) that does not allow the execution to reach to Tk event loop root.mainloop().
Using threads might help to solve this issue. You can run the data input loop in a separate thread, and run Tk event loop in the main thread as it is.
Update: I ran the code (Python 2.7.6 on Linux 3.13.0-24-generic #47-Ubuntu SMP, pyalsaaudio==0.8.2) and it was showing the data. However it could be possible that using threads on different platforms behave differently. I'm having doubts about threads in this case now, because in the comments you mentioned that even the print statement is not being called. Another approach is not to use threads and the while True loop, but to use the mainloop from Tkinter itself.
Also please not that Display.printMessage is creating a new message widget on each call. This will end up with many message widgets showing (and occupying memory). To avoid this, I'd suggest using Tkinter variables, as used on the second code example using Tkinter mainloop.
Solution 1 Using threads (minimum change to code in the question):
#!/usr/bin/python
import alsaaudio as aa
import audioop
import Tkinter as tk
import tkFont
from threading import Thread, Event
class Display():
def __init__(self, parent):
self.parent = parent
self._geom = '200x200+0+0'
parent.geometry("{0}x{1}+0+0".format(parent.winfo_screenwidth(), parent.winfo_screenheight()))
parent.overrideredirect(1)
parent.title('Listen')
parent.configure(background='#000000')
parent.displayFont = tkFont.Font(family="Unit-Bold", size=150)
def printMessage(self, messageString):
self.message = tk.Message(self.parent, text=messageString, bg="#000000", font=self.parent.displayFont, fg="#777777", justify="c")
self.message.place(relx=.5, rely=.5, anchor="c")
def setup_audio(window, stop_event):
data_in = aa.PCM(aa.PCM_CAPTURE, aa.PCM_NONBLOCK, 'hw:1')
data_in.setchannels(2)
data_in.setrate(44100)
data_in.setformat(aa.PCM_FORMAT_S16_LE)
data_in.setperiodsize(256)
while not stop_event.is_set():
# Read data from device
l,data = data_in.read()
if l:
# catch frame error
try:
max_vol=audioop.rms(data,2)
scaled_vol = max_vol//4680
print scaled_vol
window.printMessage(scaled_vol)
except audioop.error, e:
if e.message !="not a whole number of frames":
raise e
def main():
root = tk.Tk()
window = Display(root)
stop_event = Event()
audio_thread = Thread(target=setup_audio, args=[window, stop_event])
audio_thread.start()
try:
root.mainloop()
finally:
stop_event.set()
audio_thread.join()
if __name__ == '__main__':
main()
Changes:
- Moved reading sound loop into a separate function that accepts a threading Event to signal stop the loop
- Removed the
parent arg from Display.printMesssage() definition and use self.parent instead. So in the setup_autdio function when calling the printMessage(), there is no need to pass in root as an argument anymore.
Solution 2 Using Tkinter mainloop, and run all in the main thread (note that I also changed the Display.printMessage() class, mainly for performance improvements):
#!/usr/bin/python
import alsaaudio as aa
import audioop
import Tkinter as tk
import tkFont
class Display():
def __init__(self, parent):
self.parent = parent
self._geom = '200x200+0+0'
parent.geometry("{0}x{1}+0+0".format(parent.winfo_screenwidth(), parent.winfo_screenheight()))
parent.overrideredirect(1)
parent.title('Listen')
parent.configure(background='#000000')
parent.displayFont = tkFont.Font(family="Unit-Bold", size=150)
# using a Tkinter variable to display different messages, on the same widget
self.messageVar = tk.StringVar()
self.message = tk.Message(self.parent, textvar=self.messageVar, bg="#000000", font=self.parent.displayFont, fg="#777777", justify="c")
self.message.place(relx=.5, rely=.5, anchor="c")
def printMessage(self, messageString):
self.messageVar.set(messageString)
def update_ui_from_audio(data_in, window):
l, data = data_in.read()
if l:
# catch frame error
try:
max_vol=audioop.rms(data,2)
scaled_vol = max_vol//4680
print scaled_vol
window.printMessage(scaled_vol)
except audioop.error, e:
if e.message !="not a whole number of frames":
raise e
def main():
root = tk.Tk()
window = Display(root)
# Set up audio
data_in = aa.PCM(aa.PCM_CAPTURE, aa.PCM_NONBLOCK, 'hw:1')
data_in.setchannels(2)
data_in.setrate(44100)
data_in.setformat(aa.PCM_FORMAT_S16_LE)
data_in.setperiodsize(256)
def audio_read_loop():
update_ui_from_audio(data_in, window)
root.after(1, audio_read_loop)
root.after(1, audio_read_loop)
root.mainloop()
if __name__ == '__main__':
main()
Changes:
- Move the code that reads data from PCM device, and updates the UI into a separate function named
update_ui_from_audio
Display.printMessage uses a StringVar to update the message, so only 1 message widget is used.- Replace the
while True loop, with another function that calls the update_ui_from_audio function, and asks Tkinter to call it again in a milisecond (hence, acting like a loop).
