This is an interactive demo of "Flite+hts_engine" speech synthesizer which is software capable of making artificial speech in place of a real human.

Speaker	slt (Female)
Transformation	(-0.8 -- 0.8, Default: 0.55)
Pitch-shift	(-24 -- 24, Default: 0)
Speech rate	(0.5 -- 2.0, Default: 1.0)
Input text (Max number of letter = 200)

Synthesis result

Samples

• "Not at this particular case, Tom, apologized Whittemore." (Transformation: 0.55 Pitch-shift: 0 Speech rate: 1.0) wav
  
• "For the twentieth time that evening the two men shook hands." (Transformation: 0.6 Pitch-shift: -9 Speech rate: 0.9) wav
  

Options

• The voice characteristics becomes feminine if "Transformation" value is set lower and it becomes masculine if "Transformation" value is set higher.
• "Pitch-shift" value is used to control the pitch of synthesized speech in a half-tone.
• The rate of speech becomes fast if "Speech rate" value is set higher and it becomes slow if "Speech rate" value is set lower.

Input text

• The maximum number of letter for speech synthesis is 200.

Flite+hts_engine SPTK

HTS STRAIGHT

• Dec. 25, 2012 [Ver. 1.5]
  Update Flite+hts_engine to version 1.04.
  Update acoustic model.
• Dec. 25, 2011 [Ver. 1.4]
  Update Flite+hts_engine to version 1.03.
• Jul. 7, 2011 [Ver. 1.3]
  Update Flite+hts_engine to version 1.02.
  Update acoustic model.
• Dec. 25, 2010 [Ver. 1.2]
  Update Flite+hts_engine to version 1.01.
  Change sampling rate from 32kHz to 48kHz.
• Jul. 26, 2010 [Ver. 1.1]
  Fix a tiny bug in text analysis.
• May 14, 2010 [Ver. 1.0]
  Release.