Groq Whisper: How to Create Podcast Chat Application?
Вставка
- Опубліковано 26 чер 2024
- In this video, we'll dive into an exciting project where you can convert your podcasts into a dynamic chat application. Using Groq's Whisper for speech-to-text and Pine Cone for data storage, we’ll guide you through every step to create a powerful Q&A chat app. Whether you have one podcast or hundreds, this tutorial will show you how to manage them efficiently!
🚀 What You’ll Learn:
How to split podcast audio into manageable chunks.
Using Groq's Whisper for accurate speech-to-text conversion.
Storing and querying data with Pine Cone.
Building a user-friendly interface using Streamlit.
💻 Tech Stack:
Groq & Whisper: For converting audio to text.
Pine Cone: For storing and querying text embeddings.
Streamlit: For creating an interactive user interface.
🔧 Setup Instructions:
Create a Pine Cone Account and generate your API key.
Set Up Your Environment: Install necessary Python packages (Groq, Langchain, Streamlit).
Prepare Your Audio Files: Organize your MP3 files into chunks.
Run the Application: Follow our coding steps to see your podcast chat app in action.
🔗 Links:
Patreon: / mervinpraison
Ko-fi: ko-fi.com/mervinpraison
Discord: / discord
Twitter / X : / mervinpraison
Sponsor a Video or Do a Demo of Your Product: mer.vin/contact/
Code: mer.vin/2024/05/groq-whisper-...
📢 Stay Connected:
Like & Share: If you found this tutorial helpful.
Comment: Let us know if you have any questions or need further assistance.
📅 Timestamps:
0:00 - Introduction and Overview
0:35 - Setting Up the Environment
1:45 - Speech to Text Conversion with Whisper
3:15 - Creating and Querying Embeddings
4:55 - Building the Basic App
6:30 - Advanced Features: Multiple MP3 Files
8:10 - Storing Data in Pinecone
10:00 - Creating the User Interface
12:30 - Running and Testing the App
#SpeechToText #Fastest #PodCast #App #Groq - Навчання та стиль
This is amazing 🤩
Sweet 🤯
Another awesome tutorial
Can you try to make a real time chatbot and see how close it can get to respond time of GPT-4o
This is Amazing~
Great Vid! Parameterising everything I. Passing every thing even the audio file would have been amazing. Nevertheless looking forward to a video like that !
Now expand on that to Stream Conversations to LLM + Stream the LLM's Text to Voice + allow for interruptions.
Use a good female voice, and you have "HER" AI like the Movie.
You can do IT❗❗❗ and you can truly say "This is Amazing" ❗
This is really amazing, can u please add voice streaming?
What does voice streaming mean
@@seththunder2077 realtime audio transcription (record and transcribe in real time)
@@seththunder2077 Instead of the whole conversation given as an MP3 to the LLM, little segments of your voice stream to the AI as you say them.
How do we access the private beta?
Getting the following error -
message: 'The model `whisper-large-v3` does not exist or you do not have access to it.',
Can you create a speech to speech real time one?
Why not use chroma db?
How did you get access to the whisper model? I keep getting a model does not exists?
yes i got the same issue
Tell about Autocoder 33b
I think we don't have access to qroq whisper model
Yeah I wasn't able to have access either. I wonder how people are getting access?
thats so old technique, i used already 2 month ago. come up with something new dude.
Do you have yr new technique on UA-cam? Pls provide link
Hi Mervin.
First of all, I would like to pay you a huge compliment. I've been following you for a few weeks now and I think it's fantastic what you're doing and that you're sharing the code immediately.
Unfortunately, I'm encountering an error with this project and am unsure how to proceed. Perhaps you have an idea and could kindly assist me?
The Error code: 413 - {'error': {'message': 'Request Entity Too Large', 'type': 'invalid_request_error', 'code': 'request_too_large'}}
Groq Whisper API docs : "File uploads are limited to 25 MB. The following input file types are supported: mp3, mp4, mpeg, mpga, m4a, wav, and webm."
Your file must be too large, got the same issue at first ! ;)