Chat with us, powered by LiveChatAssemblyAI Speech-to-Text API | Automatic Speech Recognition
See what’s new

Automatic speech recognition.
Incredible accuracy.

AssemblyAI’s Speech-to-Text API is trusted by Fortune 500s, startups, and thousands of developers around the world. Accurately transcribe audio and video files with a simple API. Extract insights like topics, sentiment, and more. Powered by advanced AI.

Start now for free
Built for developers
No credit card required
Integrate in seconds
A console-like illustration, showing the JSON response of AssemblyAI's API requestA terminal-like illustration, showing AssemblyAI's API request

AssemblyAI’s Speech-to-Text APIs are trusted by companies of every size – from startups to Fortune 500s.Read more

Start building with Speech-to-Text today


Sign up for free. Get a free API token, and integrate into your code in seconds. No credit card required.

Start now

Start transcribing. Automatically transcribe audio and video files with high accuracy. AssemblyAI's Speech-to-Text API is powered by advanced AI research.

Read our API docs

Understand your data. Automatically extract key insights from your data like Topics, Sentiment, Sensitive Content, and more.

View all features

Powerful features
Simple API.

View all features

Automatic Transcription. Convert audio and video files into text in seconds with human level accuracy. AssemblyAI's Speech-to-Text is powered by advanced AI models.

Speaker Diarization. Automatically detect the number of speakers in an audio or video file, and "who spoke when".

PII Redaction. Detect and redact sensitive PII information like credit card numbers, names, and medical injuries from transcripts.

Real-Time Streaming Transcription. Low latency real-time streaming speech recognition, with pinpoint accuracy, over WebSockets.

Topic Detection. Built on top of the IAB taxonomy, detect the topics spoken in your audio and video content.

Summarization. Automatic time-coded summaries of what's being spoken in your audio and video content.


1,321,928. The average number of files transcribed with AssemblyAI's Speech-to-Text APIs every single day.

About our customers


99.99%+ uptime. Thousands of companies - from startups to Fortune 500s - trust AssemblyAI's Speech-to-Text API for critical workloads in production.

Read our documentation


Advanced AI. Powered by advanced deep learning models, AssemblyAI’s Speech Recognition and NLP models provide State-of-the-Art results.

View all features

We're a deep learning company
Backed by top investors like Y Combinator and organizations like NVIDIA.

Our team in a Zoom meeting. There are twelve squares, and in each one of them there is a team member.AssemblyAI's team in one of its offsites. There are ten people in the photo; five of them are sitting down. Interestingly enough, eight are wearing black shirts.

Join our team. We're a fully remote team of researchers and engineers. Our team includes top AI researchers with 20+ years of experience in Machine Learning, Speech Recognition, and NLP from places like Amazon, Cisco, Apple, BMW, and MyCroft. Help us research advanced deep learning models, and give developers access to great AI tech!

We're hiring

See how AssemblyAI's Speech-to-Text APIs perform

Start building today
No credit card required.

Start now for free
Built for developers
No credit card required
Transcribe in minutes