Article read out

[arvindpdmn]

Using text-to-speech, read out the article. Possible options:

• Control voice: male or female
• Speed of reading
• Read full article or only current selection of text

Delivering all articles in this form as a podcast is not within the scope of this work. It will be a separate task.

[arvindpdmn]

Resources:

• https://github.com/ken107/read-aloud
• Firefox source code: look at Firefox Reading Mode (doesn't support reading of selected text)
• Adobe Acrobat also has this feature but code may not be open
• NVIDIA demo code: https://colab.research.google.com/github/NVIDIA/NeMo/blob/r1.0.0b3/tutorials/NeMo_voice_swap_app.ipynb
• https://github.com/TensorSpeech/TensorFlowTTS

[arvindpdmn]

Code branch iss507 created but postponing to a future release:

• using SiteSpeaker, depends on external site
• doesn't refresh audio as soon as when content changes: looks like a long cache is maintained
• question not read out: consider changing it to h3 tag and updated styling