Introduction

This library provides simple Pythonic interface to use specified text-to-speech synthesizer.

It supports following synthesizers:

• macOS's builtin synthesizer
  • Only allows personal, non-commercial use (as stated in macOS software license agreement).
  • Outputs mp4 file.
• eSpeak NG
  • Requires installing eSpeak NG CLI first.
  • Outputs wav file.
• Amazon Polly
  • Requires installing with extra simple-talk[polly]
  • Must specify engine and voice
  • Outputs mp3 file.

By default, it automatically normalizes audio volume and outputs mp3 file.

How to Install

It is available on PyPI under name simple-talk.

You also need to install Spacy model by pip install $(spacy info en_core_web_trf --url).

How to Use

It is straight-forward as below:

In [1]: from simple_talk import SimpleTalk

In [2]: s = SimpleTalk()

In [3]: s.talk("Hello world!", "output")

In addition, you can specify voice and/or synthesizer when constructing SimpleTalk object.

You can specify output filename without suffix, and it will return you full output filename with respective suffix. Otherwise, if output filename is not specified, an object of class AudioFileClip from MoviePy is returned, and you can further use it to write to any audio file with any MoviePy supported audio format.

...

In [3]: clip = s.talk("Hello world!")

In [4]: clip.write_audiofile("output.mp3")