Тёмный
LTPhen von Ulife
LTPhen von Ulife
LTPhen von Ulife
Подписаться
What can you do with GPT-4?
7:07
Год назад
YOU ARE NOT READY
4:55
Год назад
Комментарии
@marscitizen8729
@marscitizen8729 18 минут назад
🎉
@mayoutoobid
@mayoutoobid 15 дней назад
I know gTTS can be run locally without API key. Can the other 4 do the same?
@ltphenvonulife
@ltphenvonulife 14 дней назад
Some of them yes
@Carl-upi
@Carl-upi 25 дней назад
Cool edit
@glamgar
@glamgar 25 дней назад
For me Claude is better but so limited :/
@ltphenvonulife
@ltphenvonulife 25 дней назад
Yes. I think they will improve with upcoming models
@afriendofafriend223
@afriendofafriend223 Месяц назад
This is not GPT 5, It is GPT 01-preview. Quit it with the clickbait.
@jackfrost6268
@jackfrost6268 Месяц назад
GPT5?? ........
@Ryguy12543
@Ryguy12543 Месяц назад
You are using o1-preview and o1-mini. There is no o1-omni (GPT-4o and 4o-mini are the Omni models).
@rolivasarhelyi
@rolivasarhelyi Месяц назад
Proof that it is 1.5M?
@ltphenvonulife
@ltphenvonulife Месяц назад
x.com/levelsio?t=0aJQUuodZjYwZqUZ-u2ZIg&s=09
@wilsman77
@wilsman77 Месяц назад
i'd say its more like 1.4M :(
@ltphenvonulife
@ltphenvonulife Месяц назад
@@wilsman77 Latest update: PhotoAI.com $148K/m and 148k$*12 = 1.776.000 $ / year We can not really prove it but the usecase is convincing, so I will say it is surely making that amount of money
@rolivasarhelyi
@rolivasarhelyi Месяц назад
@@wilsman77 i'd say its a clickbait :S
@animationgaming8539
@animationgaming8539 Месяц назад
nice video
@ltphenvonulife
@ltphenvonulife Месяц назад
Thanks
@yantech.4249
@yantech.4249 Месяц назад
Big thanks for this tutorial. Can you make more step-by-step project ?
@ltphenvonulife
@ltphenvonulife Месяц назад
Yeah. Sure
@Antel_007
@Antel_007 Месяц назад
As someone who speaks 6 languages myself. I find this app idea very fascinating to look at. would definitely be studying this. and hopefully if there are any collaboration opportunities in the future then it would be a dream come true actually.
@ManishKumar-rz9ub
@ManishKumar-rz9ub Месяц назад
I am getting this error while trying to run the mpi. HYDU_create_process (lib/utils/launch.c:73): execvp error on file python (No such file or directory) I am mac OS.
@ltphenvonulife
@ltphenvonulife Месяц назад
Hello. It will be hard to debug with only that. Contact me on instagram, or discord I will help you fix it
@laughingbrick4824
@laughingbrick4824 Месяц назад
Fuckin Open AI is not free whyyyy 😭
@ltphenvonulife
@ltphenvonulife Месяц назад
😂😂😂sorry @laughingbrick4824. Maybe one day AI will be free 😉
@asketsystem9461
@asketsystem9461 2 месяца назад
Great Job, please did you provide the Dockerfile ? Having issue while running docker desktop
@ltphenvonulife
@ltphenvonulife 2 месяца назад
Hello. Yes of course. I will add it to the github
@vivekgangurde9685
@vivekgangurde9685 3 месяца назад
can we able to deploy this application on aws ?
@ltphenvonulife
@ltphenvonulife 3 месяца назад
Yes sure. You can use aws EC2 service for that
@matthewferguson5464
@matthewferguson5464 3 месяца назад
Great work man
@GenteItalianaOficial
@GenteItalianaOficial 3 месяца назад
hello, do you offer the service to create this system? thank you
@ltphenvonulife
@ltphenvonulife 3 месяца назад
Hello, yes we offer AI services. You can visit launchpad.ulife.ai to have an overview. Or you can contact me on linkedin using the social links in the description of the video
@pawekowalski510
@pawekowalski510 4 месяца назад
I want to do exactly the same simple website like Martha. Could you help me?
@ltphenvonulife
@ltphenvonulife 4 месяца назад
Hei. Yes sure. What do you need ?
@pawekowalski510
@pawekowalski510 4 месяца назад
@@ltphenvonulife Thank you, could I e-mail you?
@Charles-me7vf
@Charles-me7vf 4 месяца назад
How and what transforms the code into the tool it runs where could you tell me
@ltphenvonulife
@ltphenvonulife 4 месяца назад
Hi. I don't understand your question
@Charles-me7vf
@Charles-me7vf 4 месяца назад
Hello, when I try to access the website and generate the recap error audio, the host is not in the hosting name list, how can I solve this in the code
@Charles-me7vf
@Charles-me7vf 4 месяца назад
Recaptcha error
@ltphenvonulife
@ltphenvonulife 4 месяца назад
You need to add your host in the captcha whitelist in the dashboard
@MxttyIce
@MxttyIce 4 месяца назад
can you use this to finetune? or is this just to run what is given?
@ltphenvonulife
@ltphenvonulife 4 месяца назад
Hello. Yes you could fine-tune that model
@shubhambhushan338
@shubhambhushan338 5 месяцев назад
is there any limit or it is unlimited?
@ltphenvonulife
@ltphenvonulife 5 месяцев назад
It depends on how much you pay
@harrisonekpobimi4764
@harrisonekpobimi4764 5 месяцев назад
Wonderful ideas. i would love to see a tutorial on how elearning platform can leverage AI
@englishplease8726
@englishplease8726 5 месяцев назад
how to learn English with Indonesian language as mother tongue in this AI apps ( your creation AI language apps), any code should I change?....please kindly answer...thank you in advance
@ltphenvonulife
@ltphenvonulife 5 месяцев назад
Hello. You should not change that much. Nor Aly you should find an LLM that can speak that language, maybe you can find it on hugging face and then you can simply change the model and the prompt to just teach hindi
@melaneemelanee4205
@melaneemelanee4205 5 месяцев назад
where we van find the source code?
@ltphenvonulife
@ltphenvonulife 5 месяцев назад
Hello. It was only some test. I did not keep the source code
@akankzltas1510
@akankzltas1510 6 месяцев назад
from TTS.utils.manage import ModelManager from TTS.utils.synthesizer import Synthesizer import sounddevice as sd import soundfile as sf def play_audio(filename): # Ses dosyasını oku data, fs = sf.read(filename, dtype='float32') # Ses dosyasını oynat sd.play(data, fs) # Oynatma işlemi tamamlanana kadar bekle status = sd.wait() # Model yöneticisi oluştur path = "/home/Akon/PycharmProjects/pythonProject/.venv/lib64/python3.10/site-packages/TTS/.models.json" model_manager = ModelManager(path) # TTS modelini indir ve yükle model_path, config_path, model_item = model_manager.download_model("tts_models/en/ljspeech/tacotron2-DDC") # Sentezleyiciyi oluştur synth = Synthesizer( tts_checkpoint=model_path, tts_config_path=config_path, use_cuda=False # GPU kullanmak istiyorsanız True yapın ) # Metni sentezle text = "Hello my name is Akon,I'm a computer programmer so I'm verry happy" audio = synth.tts(text=text) # Ses dalgasını dosyaya kaydet output_path = "output.wav" synth.save_wav(audio, output_path) # Kaydedilen ses dosyasını dinle play_audio(output_path)
@Nursultan-sz9ih
@Nursultan-sz9ih 6 месяцев назад
why this doesn't work this already open but when i write text it's couldn't read
@ltphenvonulife
@ltphenvonulife 6 месяцев назад
Hello. Which error are you getting in the console
@USAHeavenlyTouchMinistry
@USAHeavenlyTouchMinistry 6 месяцев назад
What is that program you are using to type commands and code? Thanks in advance.
@ltphenvonulife
@ltphenvonulife 6 месяцев назад
Hello. It is vscode
@raghavgupta-f6x
@raghavgupta-f6x 6 месяцев назад
Is this app live somewhere?
@ltphenvonulife
@ltphenvonulife 6 месяцев назад
Hello. There is a last part of the tutorial about deployment but it may takes more time. The final.code os in the description and you can directly deploy it
@akj3344
@akj3344 6 месяцев назад
Please provide github link as well. Kind of hard to debug my code
@ltphenvonulife
@ltphenvonulife 6 месяцев назад
Hello 👋. The description had been updated with the source code
@akj3344
@akj3344 6 месяцев назад
@@ltphenvonulife Thank you so much mate. Liked and subbed!
@rohitkochikkatfrancis
@rohitkochikkatfrancis 6 месяцев назад
WHOA!!! THIS IS CRAZY !!! Can't see videos that combines latest techs like nextjs, clerk and AI !!! Really amazing video 😍
@Aman_yadav1419
@Aman_yadav1419 6 месяцев назад
Amzing ... Keep posting content like this ❤
@ltphenvonulife
@ltphenvonulife 6 месяцев назад
Thanks 🙂
@joshbleijenberg4000
@joshbleijenberg4000 6 месяцев назад
This is amazing !!
@nisujadhav2764
@nisujadhav2764 6 месяцев назад
I got mpv not found error can you solve ? On widows
@ltphenvonulife
@ltphenvonulife 6 месяцев назад
Hello. You can use this link to download the windows version of mpv: mpv.io/installation/
@yantech.4249
@yantech.4249 6 месяцев назад
Great project and big thanks for your work Sir.
@ltphenvonulife
@ltphenvonulife 6 месяцев назад
My pleasure
@ltphenvonulife
@ltphenvonulife 6 месяцев назад
My pleasure
@datasciencetoday7127
@datasciencetoday7127 7 месяцев назад
any chance you can upload this in better video quality. it's 360p now
@ltphenvonulife
@ltphenvonulife 7 месяцев назад
Hello. It is still Processing the HD vidéo. It Will be ready in few hours
@ltphenvonulife
@ltphenvonulife 7 месяцев назад
@datasciencetoday7127 The quality is better now
@datasciencetoday7127
@datasciencetoday7127 7 месяцев назад
thank you @@ltphenvonulife
@mariron42
@mariron42 7 месяцев назад
How use this in spanis?
@ltphenvonulife
@ltphenvonulife 7 месяцев назад
Hello. It is possible using one of the voices in their model lists
@mariron42
@mariron42 7 месяцев назад
@@ltphenvonulife Thanks, it seems there is only one good voice in Spanish, maybe I have to train another model
@ltphenvonulife
@ltphenvonulife 7 месяцев назад
@@mariron42 ohhh yes. That's right unfortunately
@FranklynKogolo
@FranklynKogolo 7 месяцев назад
Do I delete the NEXT_PUBLIC_API_URL= "localhost:8000" in env and store my api key?
@bobhillier921
@bobhillier921 7 месяцев назад
I would watch this but the screen is impossibly too small to see. I don’t like to complain but….
@FranklynKogolo
@FranklynKogolo 7 месяцев назад
where do I put my api keys?
@ltphenvonulife
@ltphenvonulife 7 месяцев назад
Hello. Create a .env file and add the env variables there
@FranklynKogolo
@FranklynKogolo 7 месяцев назад
@@ltphenvonulife Thanks, after that do I make any changes in the next.config.js file?
@ltphenvonulife
@ltphenvonulife 7 месяцев назад
No need. It will work directly
@FranklynKogolo
@FranklynKogolo 7 месяцев назад
Ok, do I delete the NEXT_PUBLIC_API_URL= "localhost:8000" in the .env file you gave?@@ltphenvonulife
@mindspacezone3445
@mindspacezone3445 7 месяцев назад
Hi, what file contains the mic icon image? I am new to this and I am trying to modify this demo but I don't know where to find the url.
@ltphenvonulife
@ltphenvonulife 7 месяцев назад
Hello. It is an svg Icon. It is directly in the code. You can directly change It by copying the content of an svg It replace It by an image Tag
@AtomicPixels
@AtomicPixels 7 месяцев назад
I’m worried people will take this example as a viable solution for that space. Chatting about a specific subject such as languages doesn’t make this a product. It’s still yet another chat UI using already m problems solved a thousand times over already by NLP models, yet still attempting to cash out on what’s basically a prompt and some UI libraries.
@ltphenvonulife
@ltphenvonulife 7 месяцев назад
Hello @maskedvillainai. Thanks for your response. My goal with this tutorial is not to build the next chatgpt but instead to show how people can leverage open-source models and close source models(llm and tts) to build complete solutions. It is nice to have complex tools with intense, complex, internally built AI, but we always need to start with the basics. The tutorial is the basics. The need to make it a product/company is an iterative process where you build your MVP, which can be that simple chat and then iterate progressively with the users' feedback.
@ltphenvonulife
@ltphenvonulife 7 месяцев назад
Let me know if you have usecases people actually need that fine-tuning can solve, I will be happy to do a tutorial about it. 😊
@fantoSama
@fantoSama 7 месяцев назад
Amazing,thanks for sharing. I am looking forward to the next part
@ltphenvonulife
@ltphenvonulife 7 месяцев назад
Thanks a lot @fontoSama
@justinwall3791
@justinwall3791 7 месяцев назад
🎉great content. I'm thinking about using it for my team. Github co-pilot seems intrusive
@Hgyhgyggt
@Hgyhgyggt 7 месяцев назад
Hey man! I am having some issues getting this to run. Any way I can contact you on discord or something to get some help? Very interesting project. Could really help my business. Also, what would you reccomend for making outbound calls? Twillo, or some other alternative? Is there anything open source I could connect to make calls with my phone perhaps? Thanks man!
@ltphenvonulife
@ltphenvonulife 7 месяцев назад
Hello. Yes sure. You can contact me via discord, or write to ulifeai on linkedin, or whatever works for you. We will schedule a meeting and review your project.
@Hgyhgyggt
@Hgyhgyggt 7 месяцев назад
@@ltphenvonulifeI really appriciate that, thank you very very much! Will be adding you :D
@himanshi191
@himanshi191 7 месяцев назад
how we can make fronted part
@ltphenvonulife
@ltphenvonulife 7 месяцев назад
Hello. You can checkout this video on my channel where we build the front-end ru-vid.com/group/PLQKQxPyoZNBBqNhruSD5TqMOUySxV8BsW&si=d1X6A30vz1sx2xrp
@JdjdjJjj-pg2zb
@JdjdjJjj-pg2zb 8 месяцев назад
You're gonna be famous good luck bro
@ltphenvonulife
@ltphenvonulife 8 месяцев назад
Thanks a lot man
@MarcoRonco
@MarcoRonco 8 месяцев назад
Thanks for the tutorial! ^_^ Is there a way to have Gemma access the web to read content of web pages? Thanks :)
@ltphenvonulife
@ltphenvonulife 8 месяцев назад
Yes just use #before your url. Like #www.google.com
@MarcoRonco
@MarcoRonco 7 месяцев назад
@@ltphenvonulife thanks but it doesn't seem to work. I wrote: "please visit #google.it and search for "gatto" and print the first 3 results" Result Gemma:7b Explanation of the phrase "google.it gatto": Components: .... No navigation :( Am I missing something?