ozzyjohnson/tts repository overview

⁠docker-tts

Quick container for Text To Speech (TTS). Uses eSpeak with MBROLA voices and SoX to generate MP3s from snippets of text.

⁠Usage

docker run \
  -it \
  --rm \
  -v /home/ubuntu:/data ozzyjohnson/tts \
  bash -c 'espeak "The sky above the port was the color of television, tuned to a dead channel." --stdout > gibson.wav'

Will encode the provide to test to WAV.

docker run \
  -it \
  --rm \
  -v /home/ubuntu:/data ozzyjohnson/tts \
  bash -c 'espeak -f moby.txt --stdout | sox -t wav -c 1 - -t mp3 moby.mp3'

Will encode to MP3 in place using the file moby.txt as input.

Both examples are mounting a /home/ubuntu a data volume. Update the path, text and output type to suit your needs.