Building FFmpeg stream from multiple audio files in python

I want to stream images by using a python file, that come from a pipe, along with junks of ~5 second long audio files (that are then played one after another).

I tried

cmd_stream = ['ffmpeg', 
'-f', 'image2pipe', 
'-thread_queue_size', '2',
'-re',
'-i', '-', # Indicated input comes from pipe 
'-i', 'InputFiles/a%4d.wav', #use existing audio files
'-r', 25,
'-preset', 'veryfast',
'-vcodec', 'mpeg4', 
'-f' , 'mpegts',
'udp://127.0.0.1:1234']
        
pipe = sp.Popen(cmd_stream, stdin=sp.PIPE)

But this does not even start a stream, when piping images. The audio is produced only during the run, does this make a difference?