Name | Modified | Size | Downloads / Week |
---|---|---|---|
Parent folder | |||
VATSG_v1.0.4.zip | 2023-09-19 | 469.9 MB | |
VATSG_v1.0.4.z01 | 2023-09-18 | 2.1 GB | |
VATSG_v1.0.4_setup.exe | 2023-09-18 | 1.4 GB | |
README.md | 2023-09-18 | 871 Bytes | |
VATSG v1.0.4.tar.gz | 2023-09-18 | 29.0 MB | |
VATSG v1.0.4.zip | 2023-09-18 | 29.0 MB | |
Totals: 6 Items | 4.0 GB | 56 |
(ENG) - There was a error when if mp3 file is too long and text is too much, then error occured. And also if mp3 file is short enough to get no transcription or has only voice which is sobbing, it also made error. This version, I modified to solve this problem. - I changed to stable-ts(https://github.com/jianfch/stable-ts) to make more reliable and more accurate transcription. It detect audio by word level. So It will be more accurate.
(KOR) - 영상이나 음성이 너무 길어서 텍스트양이 많은 경우나, 너무 짧거나 단음영상이라 아무런 텍스트가 생성이 안되는 경우에 에러가 났었는데 이를 해결했습니다. - stable-ts (https://github.com/jianfch/stable-ts) 를 도입했습니다. 단어 단위로 더 정확한 필사를 보여줍니다. 보다 정확한 audio detection을 체감할 수 있을 겁니다.