Hey 👋, this weekly update contains the latest info on our new product features, tutorials, and our community.
Get More with Conversational Intelligence
With the vast sea of digital conversational data generated daily—from virtual meetings and call centers to chatbots—enterprises are increasingly turning to AI to navigate this information flood. One exciting and fast-growing solution is Conversational Intelligence AI, which is swiftly rising to meet this challenge.
AssemblyAI offers many capabilities to draw the most insights from your audio data. Some of which include:
- Sentiment Analysis: Detect the sentiment of each spoken sentence in the transcript text.
- Topic Detection: Identify different topics in the transcript. The model uses the IAB Content Taxonomy.
- Auto Chapters: Summarize audio data over time into chapters to make it easy for users to navigate and find specific information.
- Key Phrases: Identify significant words and phrases in your transcript and extract the most important concepts or highlights.
Take a look at our docs to learn more.
Join the Conversation
We’re excited to see what you build using Conversational Intelligence. Share your projects and join the conversation on X.com with the hashtag #AssemblyAI and mention us @AssemblyAI.
LeMUR Improvements
Usage of Large Language Models are priced based on number of tokens used. This might be hard to keep track of for users, not knowing how many input and output tokens are used with each request.
We have added two new keys to the LeMUR response, input_tokens and output_tokens, which can help users track usage and make sure their usage stays within their threshold.
This update follows a series of improvements AssemblyAI is rolling out to help users track their usage. Recently, we rolled out spending alerts, which you can set up on your AssemblyAI dashboard to get email notifications when your balance reaches a self-determined threshold.
Fresh From Our Blog
How to Detect Hotwords with Streaming Speech-to-Text and Go: AI Assistants all have something in common, they start up with a hotword: "Hey Siri", "Alexa". In this tutorial, you'll learn how to implement a hotword detection system using AssemblyAI's Streaming Speech-to-Text API. Read more>>
Transcribe and generate subtitles for YouTube videos with Node.js: Learn how to transcribe YouTube videos and generate SRT subtitles with Node.js and AssemblyAI in this easy-to-follow guide. It uses the "youtube-dl-exec" module to retrieve information about YouTube videos and download them. Read more>>
Top Speaker Diarization Libraries and APIs in 2023: Speaker Diarization answers the question: "Who spoke when?". In this post, take a closer look at how Speaker Diarization works, why it’s useful, some of its current limitations, and the top three libraries and APIs to use. Read more>>
Our Trending YouTube Tutorials
Build an AI Voice Translator: Keep Your Voice in Any Language! (Python + Gradio Tutorial): Can you learn a new language in seconds? Probably not. But with this guide, you can learn how to build a voice translator that can translate your own voice into 30+ languages.
🚀 Master Python & Zoom API | Build a Server-to-Server App That Transcribes Recordings: In this tutorial, you will learn how to create a robust server-to-server OAuth application that automatically transcribes your Zoom recordings and saves them as text files.
Build A Talking AI with LLAMA 3 (Python tutorial): Want to talk to an AI bot? After this video, you can. This tutorial covers building a talking AI with real-time transcription, LLAMA 3 and ElevenLabs.
