# AssemblyAI

> Speech-to-text and speech understanding API platform for building Voice AI applications with industry-leading accuracy.

AssemblyAI provides industry-leading speech-to-text and speech understanding models that power Voice AI applications for thousands of companies worldwide. The platform enables developers to transcribe audio and video files, process live streaming audio, and extract insights from voice data with exceptional accuracy and low latency. AssemblyAI processes over 40 terabytes of audio daily and serves 600M+ inference calls per month.

- **Speech-to-Text** converts pre-recorded audio and video files into accurate transcripts with support for 99 languages, speaker diarization, automatic punctuation, and word-level timestamps.

- **Streaming Speech-to-Text** enables real-time transcription with ultra-low latency for voice agents and live applications, featuring built-in turn detection and unlimited concurrency.

- **Speech Understanding** provides audio intelligence capabilities including sentiment analysis, entity detection, topic detection, auto chapters, summarization, and speaker identification.

- **LLM Gateway** unifies voice-to-intelligence workflows by applying Large Language Models directly to audio content through a single API.

- **Guardrails** ensures content safety with profanity filtering, PII redaction from both text and audio, and content moderation for sensitive topics.

- **Enterprise Features** include SOC 2 Type 2, ISO 27001, HIPAA compliance with BAA, EU data residency, self-hosted deployments, and dedicated support.

To get started, sign up for a free account with $50 in credits to access the API. Install the SDK for Python, JavaScript, or other supported languages, then submit audio files or connect streaming audio to receive transcripts. The playground allows testing models without code. Enterprise customers can contact sales for volume discounts, custom deployments, and dedicated infrastructure options.

## Features
- Speech-to-Text for pre-recorded audio
- Streaming Speech-to-Text for real-time transcription
- Speaker diarization
- 99 language support
- Automatic language detection
- Sentiment analysis
- Entity detection
- Topic detection
- Auto chapters
- Summarization
- Speaker identification
- Translation
- Custom formatting
- PII redaction
- PII audio redaction
- Profanity filtering
- Content moderation
- LLM Gateway
- Word-level timestamps
- Keyterms prompting
- Custom spelling

## Integrations
AWS, Twilio, Cloudflare, Recall, LiveKit

## Platforms
WEB, API, DEVELOPER_SDK

## Pricing
Freemium — Free tier available with paid upgrades

## Links
- Website: https://www.assemblyai.com
- Documentation: https://www.assemblyai.com/docs
- EveryDev.ai: https://www.everydev.ai/tools/assemblyai