Modify demo app to generate single audio output

[stefanfrench]

Modify app.py such that at the end of the podcast generation, a single audio clip is produced (which can be downloaded) and includes all the sections from the two different speakers.

Acceptance criteria:

• When the podcast has finished generating, there should be a clickable audio clip that brings together the audio from the different speakers sections
• There should be natural sounding pauses between the speakers speaking
• Final audio output should be clearly labeled for the user

UAT:

• In the streamlit app, after clicking 'generate podcast' and after all the sections have been generated, there should be a final audio output which combines all the audio segments into one.