35 lines
1.4 KiB
Plaintext
35 lines
1.4 KiB
Plaintext
These wav files show how Sonic performs at increasing speech rates. All sound
|
|
sampels are in the public domain.
|
|
|
|
sonic.wav
|
|
This is a sonic 2X sped-up version of a public domain librivox.org recording, from
|
|
the audiobook "Princess of Mars".
|
|
|
|
soundtouch.wav
|
|
This is the same recording as sonic.wav, but sped up using soundtouch, which
|
|
uses WSOLA rather than the sonic algorithm. Even at 2X speed up, you should be
|
|
able to hear the characteristic WSOLA distortion relative to the sonic version.
|
|
|
|
talking.wav
|
|
This is my father talking, using a decent microphone and 44KHz sample rate.
|
|
|
|
talking_2x.wav
|
|
This is his voice sped up by 2X using Sonic.
|
|
|
|
espeak_s450.wav
|
|
Sonic also performs well at increasing the speed of synthesized speech.
|
|
espeak_s450.wav was generated using 'espeak -s450 -f test1.txt -w
|
|
espeak_s450.wav'. This is the highest speed currently supported by espeak,
|
|
though Sonic can speed up espeak to much faster rates.
|
|
|
|
espeak_sonic.wav
|
|
This was generated with 'espeak -f test1.txt -w out.wav;
|
|
sonic 2.6 out.wav espeak_sonic.wav'. Sonic sped it up 2.6X, which is about the
|
|
same speed as espeak at -s450. I personally feel that the sonic sped up sample
|
|
sounds better than espeak at -s450.
|
|
|
|
twosineperiods.wav
|
|
This is just two sine periods, which is too short to hear. However, it's
|
|
useful for making sure the flush function works correctly. A 2-X speedup should
|
|
result in one sine period with no distortion.
|