AssemblyAI’s Speech-to-Text APIs are trusted by companies of every size – from startups to Fortune 500.Read more➔
Submit your files. Share 5-10 audio or video files for us to build your report.➔
We build a custom report. We'll run your files through our API along with any other API you're looking into.
Review custom report. We'll assemble the results for you to dig into and reproduce using source files and code.
Is the custom report free? Yes!
How is accuracy calculated? Word Error Rate (WER) is the industry standard for calculating the accuracy of an automatic speech recognition system. The WER compares the predicted transcription to a human transcription for an audio file, and counts the number of insertions, deletions, and substitutions made by the automatic speech recognition system in order to derive the WER. TLDR - we compare API transcripts with human transcripts to calculate Word Error Rate (WER).
Why should I get a custom report? Shopping for speech-to-text can be tricky. Every provider claims they have the "best" accuracy. Instead of promising, we'd like to show you first hand how well we perform. We provide all raw transcripts in case you'd like to run any additional internal benchmarking.