Linux Text To Speech Comparison: Flite Vs Pico2Wave Vs Festival

Comparison of open-source text to speech (TTS) software in terms of pleasantness, comprehensibility, and modularity.

Text 2 Speech on Android and Web (Update)

I love audiobooks, because they can utilize brain-time, which would otherwise be lost e.g. time when walking, running or hiking. Problem of audiobooks however used to be their scarcity, which I always struggled with especially in case of technical books. I used to generate the books to audio files. Since then may real-time text-to-speech software is available, especially on my mobile device where it matters the most:

I mostly play text I need either via:

Linux Tools for Text-to-Speech

Flite in comparison to Festival has:

  • pleasant voices
  • samely comprehensible voices
  • big problems when combination with ffmpeg streaming encoding over bash pipes (This is why I don’t use it for audiobook generation.)
  • designed for embedded devices (see docs )

pico2wave in comparison to Festival has:

  • similarly pleasant voice
  • less comprehensible voice
  • problems when one needs to be piping text into it and resolvable problems when one needs to be piping from it: ln -sf /dev/fd/1 /tmp/stdout.wav; while IFS= read -r line; do pico2wave -w /tmp/stdout.wav "$line"; done | ...
  • designed for embedded devices

Audiobook Generation

I’ve put together a simplistic tool for generating audiobooks from plain text that works! I call it text2gsm as it converts plain text file into GSM WAV compression format optimal for voice.

I used above scripts to compress Hammings lectures for download - read more here.

Created on 26 Dec 2016.
Thank you

About Vaclav Kosar How many days left in this quarter? Twitter Bullet Points to Copy & Paste Averaging Stopwatch Privacy Policy
Copyright © Vaclav Kosar. All rights reserved. Not investment, financial, medical, or any other advice. No guarantee of information accuracy.