Video

Great to see you

Are you facing any issue accessing the website ?

@@AnilChandraNaiduMatcha I left an inquiry on the website, but I didn't hear any response.

@@user-ks3le3sd4j Can you share the issue to support@vadoo.tv

@@AnilChandraNaiduMatcha There is no issue, I just hope to run your service in Korea myself.

@@user-ks3le3sd4j Ok, we just replied back. Let us know

Dude!!!!!! please This sounds eFing Amazing

Yes, 1 video is free to generate. After that has to go a paid plan to generate more videos

Have used openai for ease of use. But it can easily be replaced with any other LLM such has Llama by using together with Ollama api

Yes

Great to connect

Sure, will add a colab for easy run

You can connect with me on Twitter x.com/matchaman11

the prompt plays fine for first time and after that it throws an exception on 2nd question Speech Final: I need to, make a table reservation. ChatCompletion(id='chatcmpl-9rpjatLaXsaQ5w25oTuBnR1r4j622', choices=[Choice(finish_reason='stop', index=0, logprobs=None, message=ChatCompletionMessage(content='Of course, I can help you with that! When would you like to make the reservation for, and how many people will be joining you?', role='assistant', function_call=None, tool_calls=None))], created=1722617278, model='gpt-3.5-turbo-0125', object='chat.completion', service_tier=None, system_fingerprint=None, usage=CompletionUsage(completion_tokens=29, prompt_tokens=989, total_tokens=1018)) Exception in LiveClient._listening: [Errno 13] Permission denied: 'output_audio.mp3' Exception in LiveClient._listening: [Errno 13] Permission denied: 'output_audio.mp3' Handled Error: { "description": "Exception in LiveClient._listening", "message": "[Errno 13] Permission denied: 'output_audio.mp3'", "type": "Exception", "variant": "" } Connection Closed Traceback (most recent call last):

Please mention your error and raise an issue in Github

Yes, gpt-4o-mini is supported by Assistants api

Definitely you will be able to use it. It should perform decent on a printed statement. If there is some handwritten text, it might have some issues

stackoverflow.com/questions/46986398/import-data-into-google-colaboratory You can find the way to upload a file to google colab here

stackoverflow.com/questions/46986398/import-data-into-google-colaboratory You can find the way to upload a file to google colab here

Sure, will try to cover it in a video

There would be no code displayed in the live version.

Currently it is using Gemini api but it can easily be replaced with a local running LLM such as Llama3 or Mistral

Yes, we can connect on Twitter

@@codingbyte847 Yes definitely, will add more options including Gemini and a local LLM in the next update

Will try to share more in a new video

@@AnilChandraNaiduMatcha That would be great any update?

@@AnilChandraNaiduMatcha would love to see that too!!!!!

@@AnilChandraNaiduMatcha Please

Thanks a lot

You can share the issue in Github, will check

You can check from console logs if there is an issue with the api key in which case you should get an invalid api key error

@@AnilChandraNaiduMatcha ah okay, i noticed in the application the reply seemed to be loading and loading... however when i went into the playground area within open ai's website, I typed a message the bot i created with the application and i got the "You've reached your usage limit" message... so i figured it was some sort of api issue, im a broke boi an open ai loves to rub it in my face

Definitely, what canva action are you looking at ?

@@AnilChandraNaiduMatcha Awesome! I would like Canva to offer graphic designs to contain the text I just generated in my customGPT. For example: I have designed a customGPT that generates lead magnet ideas for business owners. I would then like to open Canva, and incorporate that lead magnet ideas into done-for-you Canva designs. Could your tool do this?

Thanks 😄

Great idea, will add it in the next update

Great idea, will try to add it in an update

@@AnilChandraNaiduMatcha Great! I could contribute to that! Let’s get in touch!

@@pascalmariany Sure, you can start by raising an issue in GitHub

@@AnilChandraNaiduMatcha Will do! I think with an upload button you can have more reason to share this custom GPT with potential customers.

For running whisper to generate captions you might need a system with minimum 8 GB RAM. If you have a low end pc, you can replace running whisper locally with whisper api

Definitely can use a self hosted llm instead of gpt-4, will update it

Definitely, can just change the moviepy font in code and it should work

Thanks for the suggestion Srikar, will try to add it in a future video

Thanks for the support

Yes, this does the first step of identifying the viral shorts. The editing part I will discuss in a different video

Yes it's free but rate limited

@@AnilChandraNaiduMatcha thank you 👍

Thanks for the support

Thanks for the support

Definitely, Groq seems to be a revolution in AI space

No this is just to run GPTs using Assistant api, doesn't work for finetuning

@@AnilChandraNaiduMatcha Sure thanks for informing.

You can add guardrails to your instructions to prevent it from being hacked, but it is not foolproof. Another option is hide most of your functionality of your GPT behind your own api and guard it with Oauth2

@@AnilChandraNaiduMatcha ok I appreciate your insight. Great video

Any chance of hacking it so becomes uncensored?@@AnilChandraNaiduMatcha

Make sure your VerifyToken endpoint works after redirecting to OpenAI. If you wish for a No-code way of achieving this, check out gpt-auth.com/