Speech Understanding

Gain maximum value from voice data with audio intelligence models, and leverage LLM capabilities with LeMUR to extract insights, generate summaries, and more.

Use our API Contact sales

Summary

The customer, John, called Acme Corporation's customer service department to report a malfunction with his widget. Sarah, the representative, attempted troubleshooting but concluded that the widget needed repair under warranty.

Topic Detection

Customer Service, Product Support, Warranty, Troubleshooting, Repair

Key Phrases

Acme Corporation, Sarah, John, Acme Widget, malfunction, serial number, batteries, troubleshooting, repair, warranty, prepaid shipping label

Extract valuable insights from voice data

Audio Intelligence

AI models to summarize speech, redact personal information, detect hateful content, identify spoken topics, and more.

LeMUR

With a single API call, summarize meetings, generate call insights, recap action items, and more on over 100 hours of audio data.

An illustration of the AssemblyAI realtime playground. On top, there's a button with the Text "Start talking". Below, there's a timestamp and output with text "Hello today is"

Audio Intelligence

Feature-rich AI models

Summarization

Leverage our AI-powered Summarization models to automatically summarize audio/video data in your products at scale. Customize the summary types to best fit your use case.

See how in docs

Content Moderation

Detect sensitive content in your audio and video files - such as hate speech, violence, sensitive social issues, alcohol, drugs, and more.

See how in docs

Sentiment Analysis

With Sentiment Analysis, AssemblyAI can detect the sentiment of each sentence of speech spoken in your audio files.

See how in docs

Entity Detection

Identify a wide range of entities that are spoken in your audio files, such as person and company names, email addresses, dates, and locations.

See how in docs

PII Redaction

Identify and remove Personally Identifiable Information, such as phone numbers and social security numbers, from the transcription text before it is returned to you.

See how in docs

Topic Detection (IAB Classification)

Label the topics that are spoken in your audio and video files. The predicted topic labels follow the standardized IAB Taxonomy, which makes them suitable for contextual targeting.

See how in docs

Auto Chapters

Automatically generate a summary over time for audio and video files.

See how in docs

Key Phrases

Accurately identify significant words and phrases, enabling you to extract the most pertinent concepts or highlights from your audio/video file.

See how in docs

See everything in docs

LeMUR

Leverage LLM capabilities and take action on your audio data

Ask questions

Get instant answers to questions about your audio.

Create summaries

Summarize your audio data with key takeaways.

Extract data

Extract data such as topic tags from your audio to categorize and organize your audio data.

Generate content

Generate long-form or short-form written content using your audio data.

Ask questions

See how in docs

Input

What is a runners' knee?

Output

Based on the transcript, runner's knee is a condition characterized by pain behind or around the kneecap. It is caused by overuse, muscle imbalance and inadequate stretching. Symptoms include pain under or around the kneecap and pain when walking.

Explore how people are using LeMUR

Rapidly ship high-quality Generative AI features with voice data

Unifies your AI stack for audio

LeMUR is a single API that connects all of the spoken data in your application to an LLM to build generative features in your product. No need to chain multiple technologies together to go from audio file to LLM output.

Powered by AssemblyAI’s speech recognition models

High-quality LLM outputs on audio data start with quality transcriptions. LeMUR operates over AssemblyAI’s state-of-the-art speech recognition models, to ensure LLM outputs are top-notch.

Continuously updated with the latest in research

We are constantly experimenting with the latest research in LLMs and updating LeMUR with new techniques in retrieval, compression, prompt engineering, LLM performance, and more.

Launch quickly & effortlessly scale

Find product-market fit faster and ship new AI features at scale. The LeMUR API enables you to process over 200 hours of audio in a single API call, and handles over 1M tokens as input. The API is priced to scale as your audio data grows.

Continuously up-to-date and secure

Monthly updates and improvements

View weekly product and accuracy improvements in our changelog.

View changelog

Enterprise-grade security

AssemblyAI is committed to the highest standards of security practices to keep your data and your customers' data safe.

The accuracy was strong, but the great documentation and unique models like Auto Chapters and Sentiment Analysis is what really won us over.

Nathan Webb, Product Manager, Aloware

Explore more

Speech-to-Text

Build on top of the most accurate Speech-to-Text model on the market with >92.5% accuracy.

Learn more

Streaming Speech-to-Text

Transcribe audio streams synchronously with high accuracy and low latency.

Learn more

START BUILDING WITH AI

Get started in seconds

Use our API Contact sales

import assemblyai as aai
import json

transcriber = aai.Transcriber()
transcript = transcriber.transcribe(URL, config)

print(json.dumps(transcript, indent=2))

{
  "id": "6rlr37h8f4-e310-4e23-bbf3-ea5f347dc684",
  "language_code": "en_us",
  "status": "completed",
  "text": "Runner's knee is a condition characterized by pain behind or around the kneecap...",
  "confidence": 0.98122,
  "audio_duration": 3200,
  "words": [
    { "text": "Runner's", "start": 0, "end": 550, "speaker": "A", "confidence": 0.98113 },
    { "text": "knee", "start": 580, "end": 1130, "speaker": "A", "confidence": 0.95417 }
  ]
}