Datasets:

real-ssb22
/

yali-lower

Tasks:

Automatic Speech Recognition

Languages:

Size:

Tags:

speech-synthesis

License:

Dataset card Files Files and versions

63.9 MB

3 contributors

History: 25 commits

real-ssb22's picture

HuggingFace dataset + Chinese notes

72e72d7 unverified 3 months ago

wav-syllables
Improve the consonants of 'dei' and 'dao' syllables in yali-lower by amplification and by copying the best 'd' sound to the other tones, to make it clearer that a 'd' is there. Still couldn't get the 'd' of dao1 very clear though, and I'm still not happy with some of the 'den' and 'deng' ones especially 'deng1', but there doesn't seem to be enough clear source material to get that any better on yali-lower. One thing that does seem to help a tiny bit though is inserting a tiny bit of silence at the start of deng1 so it's more clearly separated from whatever the previous syllable was (e.g. de5) over 13 years ago
.gitignore

3 Bytes

gitignore over 1 year ago
LICENSE

35.1 kB

Create LICENSE about 8 years ago
Makefile

1.36 kB

HuggingFace dataset + Chinese notes 3 months ago
README.md

3.18 kB

HuggingFace dataset + Chinese notes 3 months ago
metadata.csv

113 kB

HuggingFace dataset + Chinese notes 3 months ago
setup.bat

1.03 kB

yali-lower scripts for gradint almost 15 years ago
unzipsfx.exe

49.2 kB

yali-lower scripts for gradint almost 15 years ago
yali-exe-readme.txt

1.21 kB

minor over 5 years ago