Speech-to-Text Features

Our API is built using the latest advances in Deep Learning, using cutting edge research from our in-house team of AI researchers.

Casing and punctuation of proper nouns are automatically added to the transcription text, to make transcripts produced by the API more readable.

Don't worry about file formats or sampling rates, our API supports virtually all audio/video files without any transcoding required.

Get a confidence score for each word in the transcript.

Word-by-word timestamps across the entire transcript text.

The number of speakers in the audio can be automatically detected, and each word transcribed associated with its speaker. This can help answer the question "Who Spoke When?"

Automatically detect and replace sensitive data, like credit card numbers and social security numbers, in the transcription text and source audio.

Easily export your transcription in SRT or VTT format, to be plugged into a video player for subtitles and closed captions.

The API can automatically detect key phrases and words in your transcription text - very useful for tagging content, and providing a summary of the transcription text.

Automatically detect and replace profanity in the transcription text.

We adhere to best-practice guidelines like encryption in transit and at rest.

We are not in the business of monetizing your data. Files sent to the API for transcription are never stored, and you can request the deletion of transcription text permanently from our database.

Thousands of developers and startups trust AssemblyAI to power core features in production, with over 99.9% uptime and transcript completion rate.

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.