Recent content by ironfevers

I wanted to grab the Chinese subtitle for MFYD-001 but it never got made. Time to revisit Whisper. I didn't know about the PRO version until recently. Mei2 is awesome for making it. I wanted to improve my workflow and so I made a Python script for cleaning the Japanese srt. Garbage lines of...

With an extra step of restart session, you can avoid the "numpy.dtype size changed" error without needing to upgrade those packages. The workflow is mount Google Drive, setup, click runtime>restart session, set parameters, then run Whisper. In WhisperPro, I use an initial_prompt =...

Just learned about this. Use the new model for improved translations. Change from MODEL = "accounts/fireworks/models/deepseek-v3" to MODEL = "accounts/fireworks/models/deepseek-v3-0324". My first run caused line merges at TEMPERATURE = 0.9. My new recommended default is TEMPERATURE = 0.8 Nao...

It's been a long time since I posted. I recently discovered how good Deepseek is, and I got excited. I spent ~30 hours on this Python Script to translate Chinese SRT files to English using Deepseek v3. Why use it? My Python script translates in batches, so it has the context of previous lines...

Whisper subs. I deleted a lot of hallucinations and garbage sentences, but I didn't change any pronouns. I updated the garbage list and hallucination suppression in google colab. https://drive.google.com/file/d/19oycJYPsBzwzaOUebkGVH80tty2n5KG6/view?usp=share_link. Feel free to make a copy to...

I don't see VAD (Voice Activity Detection) in that version. It was a hassle to install https://colab.research.google.com/github/ANonEntity/WhisperWithVAD/blob/main/WhisperWithVAD.ipynb locally but it was worth it. There were so many things to install like CUDA, cudnn, (I think these two are...

I also arrived at the same conclusions. After extensive testing, I've settled on 3.2 chunk_threshold and 0.3 vad_threshold for a good balance between catching short sentences and natural sounding speech. Source separation also fails to work. The spleeter package is to blame. I also use...

https://easypronunciation.com/en/merge-two-subtitle-files-online

With subs like that, use mpv player to sync them on the fly. As the video progresses and gets out of sync, I pause, then press ctrl shift left which cycles through to the previous dialogue. I don't know any other video players with this feature.

Thanks to the poster from javbus forum for updating it. Reuploaded 31000外挂字幕 https://www119.zippyshare.com/v/ccf2Ic8n/file.html zip password: www.javbus.com

There's a 29000+ Chinese subs posted on javbus forums on 2021-9-8. I reuploaded to zippyshare https://www49.zippyshare.com/v/310HZhjm/file.html Rar password: www.javbus.com Posting 7 recent Chinese subs (jul, meyd, nsfs, roe, ssis, vec) and ran it through DeepL translate.

I have an atid-467 with triple the number of lines. I can tell that most of the above subs are made with autosub or pytranscriber. Posting 9 subs translated from Chinese on DeepL: atid-467,jul-633,jul-635,jul-645,mide-944,mide-947,sprd-1422,sprd-1425,waaa-077

Nice guide. Do you have a recommended image viewer to delete blanks? It's kinda tedious to manually delete blanks. I was thinking you can click Create TXTImages which automatically detects blanks. Then you can run a script to compare to see if the RGBImage filename is in the same list as...

Got some of the recent JUL and SPRD subtitles for anyone who likes milfs. jul-540, jul-546, jul-554, sprd-1396, sprd-1397, sprd-1398, sprd-1399. All translated by Deepl with no editing. Not sure if it's mentioned but the software Everything by voidtools changed my life. It's so much faster than...