# AssemblyAI

> AssemblyAI is a San Francisco speech-AI company that builds and serves models turning audio and video into accurate text, plus higher-level 'audio intelligence' like summaries, sentiment, speaker labels, and PII redaction. Founded in 2017 by Dylan Fox, it sells a developer-first API used to add transcription, real-time streaming, and voice-agent capabilities to software. The company has raised more than $113M across seed to Series C and reports processing over a million hours of audio a day for customers ranging from startups to large enterprises.

- **Founded:** 2017
- **Headquarters:** San Francisco, United States
- **Founders:** Dylan Fox (Founder & CEO)
- **Team size:** ~92-100 employees
- **Products:** Speech-to-Text (Universal-2), Universal-Streaming, Slam-1, LeMUR, Audio Intelligence
- **Notable:** Raised $113M+ total across seed to Series C, including a $50M Series C in December 2023, Grew out of Y Combinator into a widely used speech-AI infrastructure provider, Reports processing 1M+ hours of audio per day and 600M+ inference calls per month

## Products & services

- **Speech-to-Text (Universal-2)** — Batch/pre-recorded transcription model tuned for high accuracy on meetings, media and calls, with punctuation, formatting, speaker labels and 99+ language support.
- **Universal-Streaming** — Real-time speech-to-text API delivering low-latency (~300ms) transcription for live captions and voice agents, with multilingual support added in 2025.
- **Slam-1** — A customizable Speech Language Model that combines LLM-style reasoning with audio processing to understand context, not just recognize words.
- **LeMUR** — An LLM layer over transcripts that returns summaries, chapters, sentiment, Q&A and action items in a single call, using Claude models.
- **Audio Intelligence** — A suite of models for speaker diarization, sentiment analysis, topic detection, auto chapters, entity detection and content moderation.
- **Voice Agent API** — APIs designed for building real-time conversational voice agents, including turn detection and streaming transcription.
- **Guardrails & PII Redaction** — Tools to redact personally identifiable information, filter profanity and moderate content before data leaves the pipeline.

## Achievements

- Raised $113M+ total across seed to Series C, including a $50M Series C in December 2023
- Grew out of Y Combinator into a widely used speech-AI infrastructure provider
- Reports processing 1M+ hours of audio per day and 600M+ inference calls per month
- Shipped Universal-2, Universal-Streaming and the Slam-1 speech language model
- Recognized as a High Performer in G2's EMEA Voice Recognition grid
- Added EU data residency and GDPR-compliant endpoints for European customers

## Latest updates

- **2025-10** — Released multilingual streaming, guardrails and an LLM Gateway; added newer Claude models to LeMUR.
- **2025** — Launched Universal-Streaming for real-time transcription at ~300ms latency and expanded to 99 languages.
- **2025** — Made Slam-1 and LeMUR available via an EU API endpoint for data residency compliance.
- **2024-10** — Released Universal-2, its flagship batch speech-to-text model.
- **2023-12** — Announced a $50M Series C led by Accel to build superhuman speech AI models.

## Links

- Website: https://assemblyai.com
- LinkedIn: http://www.linkedin.com/company/assemblyai
- Twitter/X: https://twitter.com/assemblyai
- GitHub: https://github.com/assemblyai
- YouTube: https://www.youtube.com/@AssemblyAI
- Facebook: https://facebook.com/AssemblyAI/

---

Profile page: https://yespress.io/assemblyai
Published by YesPress — https://yespress.io
Last updated: 2026-07-22