Transcribe up to 5 hours/month

Blazing fast transcriptions

Create custom models

Email/chat support



No usage limits

Blazing fast transcriptions

Create custom models

Email/chat support


Contact Us

Transcribing over 10,000 hours/month of audio? Contact us for volume discounts and to learn more about our private cloud offerings.

Q: Why should I choose AssemblyAI over Google/AWS?

AssemblyAI is more accurate, customizable, faster, and easier to integrate than Google/AWS. With AssemblyAI you can add thousands of custom words to the vocabulary, train custom models with your data for higher accuracy, transcribe any audio format accurately, and get up and running in minutes. Plus, we are constantly improving our technology, and use the latest and most advanced deep learning technologies to make sure you always have the most advanced and accurate Speech-to-Text when you choose AssemblyAI.

At AssemblyAI, we are 100% focused on building the best Speech-to-Text API for your company or product.

Q: How fast can I expect transcripts?

Transcriptions usually take around 20% of the audio duration to complete. For example, a 1 hour audio file will be transcribed in 12 minutes. Very short clips, less than 1 minute in length, will usually take around 20 seconds to complete.

Q: How do custom models work?

With custom models you can add support for unlimited custom vocabulary, boost accuracy for specific keywords/phrases, and boost overall accuracy. To learn more, check out the Guide on Custom Models.