ViewTube

AssemblyAI is building the best API platform for developers to transform and understand voice data with AI so that you can build amazing new products and services for the world to use.

With AssemblyAI's API, you can transcribe an audio file, get speaker labels, get a list of topics discussed, the sentiment of sentences and much more in more than 20 languages. 

AssemblyAI's most capable and highly trained speech recognition model Universal-1 is now available to use!

Universal-1 is
✅ 10% or more accurate than the next best Speech-to-text model in English, Spanish and German
👀 Reduces hallucination rate by 30% over Whisper Large-v3
🔈 Exhibits the ability to transcribe multiple languages within a single audio file
🏎 Takes 38 seconds to process a 60-minute audio file
💰 And costs as low as $0.37 per hour of audio-processed

Find out more about the model here:  https://lnkd.in/dZhPHVcW 

LLMs work wonders on text data but if you want to use audio or video files instead of text, things get a bit trickier. 

To set up a RAG application on audio data without losing valuable information (like how many people were speaking and who said what), you can include speaker labels in your transcription.

Let's see how to set this up using AssemblyAI and Haystack. 

New AI-powered tools are coming out every week if not every day. In this video, we compiled our favorite content creation tools with AI features.

With these AI tools (or creator platforms), you can plan, record, edit and publish your content with ease. They could help you write your script, remove the background of your videos, cancel the background noise, create images for your thumbnails, give you ideas for titles, dub your videos, transcribe your podcasts, add subtitles to your videos, or chop up your videos into shorter ones to share on your socials. 

Graph Neural Networks have recently made rapid strides!

Despite less hype compared to other AI buzzwords, GNNs have powered numerous achievements in the last year.

Our latest video highlights recent breakthroughs and concrete applications of GNNs. 

Andrew Huberman is a very successful podcaster. He generally talks about wellness and health in his podcast, has interesting guests over and has millions of listens on every episode. And he has a great website where you can search for keywords to find corresponding podcast episodes or parts of episodes where he discusses these topics. 

In this video, we will build the first step into making a website like Andwer Huberman's website to find podcast episodes based on keywords. You can replicate this to index any podcast that you like.! 

Catch Matt talk about Voice LLMs next week at SeedToScale. Register here:  https://lu.ma/ai-talks-12 

"In this session, Matthew Makai, VP Developer Relations & Experience, AssemblyAI will cover how to quickly integrate high accuracy speech-to-text transcription of audio and video data into new or existing code bases. Additionally, how to use a large language model (LLM) with audio and video data using only a single line of code, and building with advanced features such as content moderation, topic detection and personally identifiable information (PII) redaction." 

Step by step guide to making an AI app on audio files without coding 👇

Step 1. Make an account at assemblyai.com
Step 2. Watch our tutorial on Rivet ;) 

Unexpected skills of LLMs emerge as they get bigger. But how?
Ryan explains in our latest video. 

Thank you for your support, we are officially at a 100K! 🥳