If you press "8" at the prompt, you'll be served a fresh image from The Cat API that is rendered on-the-fly: https://github.com/jpf/dial-a-cat/blob/master/app.p

If you press "8" at the prompt, you'll be served a fresh image from The Cat API that is rendered on-the-fly: https://github.com/jpf/dial-a-cat/blob/master/app.py#L107-L133 https://github.com/jpf/dial-a-cat/blob/master/app.py#L107-L1... To get this to work, I had to serve the WAV samples to Twilio as they were generated, this is described just below the first block of code in this section of the README: https://github.com/jpf/dial-a-cat#transmitting-sstv-images-over-the-telephone https://github.com/jpf/dial-a-cat#transmitting-sstv-images-o... The reason I serve a pre-rendered audio file by default is that not every image you get from the Cat API results in an image that is recognizable after Martin M2 is done with it.