deepspeech_obs_captions

Realtime captions in obs-studio using DeepSpeech.

Installation

• Install Python 3.6 64-bit, or 32-bit depending on your OBS version.
• pip install -r ./requirements.txt
• Download trained models from https://github.com/mozilla/DeepSpeech/releases and extract them into ./deepspeech-models.

The folder structure should look something like:

• caption_stream.py
• deepspeech-models
  • lm.binary
  • output_graph.pb
  • output_graph.pbmm
  • output_graph.tflite
  • trie

OBS Setup

• Add a Text source
• Navigate to Tools > Scripts
• Set your Python path in Python Settings tab
• Add ./caption_stream.py in Scripts tab
• Select the script and assign Captions Text Source

Config

VAD Aggressiveness controls voice activation, a lower value will let more noise through.
Caption Clear Delay controls how long captions remain on the screen.
Captions Text Source controls what element captions appear on.

The other settings are probably best left alone unless you know what you are doing.
For more details, see: https://github.com/mozilla/DeepSpeech/blob/master/examples/mic_vad_streaming/README.rst#usage