Linux

A Deep Dive into Voice Cloning with SoftVC VITS and Bert-VITS2 In the previous post, I have tried a little bit of TTS Generation WebUI and found it’s interesting. So, I decide to train a usable model with my own voice. This voice cloning project explores both SVC for Voice Changing and VITS for Text-to-Speech. There is no one tool does all jobs. I have tested several tools for this project. Many of the good guides, like this, this and this, are in Chinese. So, I thought it’s useful to post my notes in English. ...

Old Stories The computer hardware used to be more playful and worth tinkering. My favorate platfom from a decade ago, Sandy Bridge on LGA 1155, can still be powerful even today. Back in the days, I used to repair people’s electronics for free. Because of that, I also received a lots of spares and e-waste in exchange. One of the best was a LGA 1155 motherboard with i5-2300 on it. Then, I bought a cheap E3-1245 and GTX 750 Ti to make it a gaming rig. I played a lot of games on that, such as Dark Souls series and Metro series. Before it was sold, last games I played on this build was Metro Exodus and Elden Ring. ...

Background Story Last week, I experienced a major data loss on my daily driver computer due to encryption failure. Last time I had encountered this sort of situation was over a decade ago with TrueCrypt on an external HDD and unfortunately lost everything that weren’t backed up. I’ve recovered everything from backup this time so I decide to document it. The entire storage on my daily driver is encrypted with LUKS and it suddenly fails on boot decryption. I don’t have the header backed up and have stopped using Timeshift for a long time due to it’s unschedulable which lead to chaotic performance impact. ...

I use Generate-subtitles as an alternative or substitute to YouTube closed captions (CC) since it does not always work as expected. When creating video contents, it comes very handy to have a high quality generated transcripts to start with. My favorite tools are Subtitle Edit and Aegisub. SE provides a great online version and works with .SRT format which fits great into my Adobe Premiere Pro workflow. It also has built-in Auto Translation and Whisper support. Aegisub works with .ASS format which fits into NixieVideoKit automated workflow. ...

No DALL-E, No Midjourney and No Colab This is a guide showing how to build your own stable diffusion server on what you already have or cheap used hardwares. It may not satisfy for a serious production use but pretty viable for learning, testing or casual use. Before we start, here’s some comments on OpenAI: The history of ChatGPT creator OpenAI, which Elon Musk helped found before parting ways and criticizing OpenAI Is Now Everything It Promised Not to Be: Corporate, Closed-Source, and For-Profit Will ChatGPT be open source? ChatGPT, how did you get here? It was a long journey through open source AI When big AI labs refuse to open source their models, the community steps in Artificial Intelligence: Last Week Tonight with John Oliver (HBO) The TRUTH about OpenAI I have been using ChatGPT and its API regulary. Since my last post, the API has been upgraded to gpt-3.5-turbo which broke my program and gpt-4 seems coming up soon. Therefore, I may not fix my code proactively. ...