The Top Rated API for Speech-to-Text

High accuracy, easy to setup, and trusted by thousands of developers and companies.

** Just in: AssemblyAI recognized as a Spring 2020 High Performer by G2 based on reviews **

Used by 1000s of developers at startups and top companies

Top-Rated Accuracy

We're a Y-Combinator funded team of Deep Learning Researchers, working towards achieving human-level accuracy with machine transcription.

Privacy Focused

We're privacy focused, and don't harvest your data for training our models like other big tech companies do.

Customizable to Your Needs

Boost accuracy for keywords and phrases, like product names, person names, or other key terms.

Integrate in Minutes

Get started in minutes with our simple REST API using any language: Python, Node, Ruby, PHP, C#, etc.

24x7 Customer Support

All customers get a dedicated Slack Channel with our engineers for 24x7 technical support and feedback.

Packed with Features

Dual channel support, punctuation, proper noun formatting, PII redaction, etc. See a full list of features here.

Ready to start building?

Contact us or sign up now.

0*3kNFqkCqG5z7C_ul.png

We're funded by the top investors in Silicon Valley including Y Combinator and TechNexus Venture Collaborative.

Comes with all the core features you need.

Top-Rated Accuracy

We apply the latest Deep Learning research to speech recognition, and release accuracy updates every 6 weeks.

Automatic Punctuation

Transcription text is automatically punctuated to make transcripts more readable.

Proper Noun Formatting

Casing is automatically added to proper nouns (like cities and names) in the transcription text.

Dual Channel Support

Transcribe dual channel/stereo audio files with the API, and get a transcript for each channel automatically.

PII Redaction

Sensitive information like credit cards and social security numbers can be automatically redacted from the transcription.

Custom Vocabularies

Easily boost accuracy for proper nouns, like person and product names, specific to your use case.

Multiple Models

Choose from a number of models that best match your data for the highest accuracy.

24x7 Customer Support

We're here to work with you as much, or as little, as you'd like. Talk to us over live chat, Slack, email, or phone.

All Audio/Video Formats Supported

The API natively supports virtually any audio/video file, so you don't need to worry about transcoding your data.

Latest from the Blog: Building an end-to-end speech recognition model in PyTorch >

Ready to get started?

Contact us or sign up now.