Skip to content
Search

Speech synthesis

lastmod 2026-04-15created 2019-01-0942 edits

Speech synthesis is a technology in which a machine automatically produces the sound waves of speech. Put simply, it records the speech of one person chosen as a model, splits it into fixed speech units, labels them with codes and feeds them into a synthesizer, and then, on instruction, recombines only the needed speech units to produce speech artificially. It is also called TTS (=Text-to-Speech).

Synthesizing individual speech segments is difficult and gives poor results, so speech synthesis is sometimes based on recordings that capture, around each segment boundary, the end of the preceding sound together with the start of the following sound. This kind of junction processing is called diphone processing.

Along with speech recognition, it is used in many areas such as translation machines and robot manufacturing technology.

Libraries

  • say converts text to audible speech using the GNUstep speech engine.
    • sudo apt-get install gnustep-gui-runtime
    • say "hello"
  • festival General multi-lingual speech synthesis system.
    • sudo apt-get install festival
    • echo "hello" | festival --tts
  • spd-say sends text-to-speech output request to speech-dispatcher
    • sudo apt-get install speech-dispatcher
    • spd-say "hello"
  • espeak is a multi-lingual software speech synthesizer.
    • sudo apt-get install espeak
    • espeak "hello"
  • ttspico
  • google-speech
  • SoX

Deep learning based

Korean TTS

See also

Favorite site

Online tools

author Zer0 sitemap