alexei_rudak

6 ms·

1.
Show HN: I open-sourced machine translation models for 12 rare languages(github.com)
6 points by alexei_rudak 2y ago|2 comments
2.
by alexei_rudak 2y ago
Thanks! But I'm still not in my dream, may be next year will be able to create technology that can do that with desired level of quality ))
3.
by alexei_rudak 2y ago
I would love to have my own service and technology to localise videos by using AI. All startups in that niche are growing so fast.
4.
by alexei_rudak 2y ago
I've trained LLM to make conversion from American English to British English and vise versa. To do that I've used 10k sentences for fine-tuning.
5.
by alexei_rudak 3y ago
I mean the case when tiny models can outperform LLM on certain tasks. For example storytelling, and only for Harry Potter world. Something very niche but with maximum quality that can't do other tasks. Just want to see an example of th
6.
Ask HN: Small Language Models instead of LLM. Does it makes sense?
3 points by alexei_rudak 3y ago|2 comments
7.
by alexei_rudak 3y ago
Tried your service. Once I sent my question and got the answer it's not possible to enter another request in the search field. Possible a bug ?
8.
by alexei_rudak 3y ago
The more and more generated content by AI (audio, images, text, files etc) will turn Internet into trash.
9.
by alexei_rudak 3y ago
The community of dancers once I decided to learn Cuban Salsa. It was the best choice for me to feel happy.
10.
by alexei_rudak 3y ago
This is a period of uncertainty in the world, so buyers are waiting for lower prices
11.
by alexei_rudak 3y ago
Local translation is a good choice in case of privacy, but small language models can't offer a good translation quality like Chrome. Brave and Vivaldi browsers use on-premise translation by Lingvanex which helps to avoid transmitting y
12.
by alexei_rudak 3y ago
Just separate all critics in 2 parts: 1) Professionals in your sphere: Take as much critics from professionals in your sphere as you can. Sometimes you need to even pay for this. Analyse and go forward. The more mistakes you will make the f
13.
by alexei_rudak 3y ago
Wrap it as iOS or Android app with ads and send to Appstores. Even with integrated ads like Admob you can easily cover your expenses
14.
by alexei_rudak 3y ago
As I know it's impossible for current large language models to return "real-time" data, because you need to retrain it after every data update. And this will take time.
15.
by alexei_rudak 3y ago
I tried several of open LM to test what they can. The performance is very poor on PC with GPU and quality is far away to use them in commercial tasks. But to play for fun and chat like a friend "Hello, what do you think about cats? &qu
16.
by alexei_rudak 4y ago
please write to alexeir@lingvanex.com
17.
Show HN: Unlimited machine translation API for $200 / Month
36 points by alexei_rudak 4y ago|35 comments
18.
I spent $1M to make my machine translator. Thoughts and Results(medium.com)
12 points by alexei_rudak 5y ago|1 comments
19.
Show HN: Free Translation API in 100 languages for your projects
1 points by alexei_rudak 6y ago|1 comments
20.
by alexei_rudak 7y ago
Right now it works such way. We use 11 languages with our translation API and Google API for others. It also depends on how much words you need to translate per request.
21.
I earned $1M with no business experience, and spent it to create a translator(medium.com)
7 points by alexei_rudak 7y ago|2 comments
22.
by alexei_rudak 7y ago
Dear community, It was very long and difficult way to success to create my translator. Please read the story below first and then ask questions about my startup and why it's worth to show. https://medium.com/@alexeiruda
23.
Show HN: Translator apps and API (Please read my story first)(lingvanex.com)
3 points by alexei_rudak 7y ago|1 comments