Inside AssemblyAI's Dictation API: Key Terms and Sub-Second Turnaround
- Source
- x.com
- Date
"um so like can you send me the uh the Q3 numbers" is what the mic heard. "Can you send me the Q3 numbers?" is what the person meant. @iHarnoorSingh walks through the Dictation API playground showing the gap close. Add your key terms, hold to record, read what comes back. Two things worth watching: The key terms field. Type in the vocabulary your users actually say (drug names, SKUs, your own product names) and those words come back spelled right instead of phonetically close. That's the part most dictation projects die on. The turnaround. Typical short clips come back in under a second, on Universal-3.5 Pro, across 19 languages.. Bring your own audio and throw the words your users get wrong at it 👇
Context
A walkthrough of the AssemblyAI Dictation API playground opens with the gap dictation tools usually leave: a mumbled 'um so like can you send me the uh the Q3 numbers' has to become 'Can you send me the Q3 numbers?' The demo shows a person holding to record and reading back what the API returns.
The specific mechanism shown is a key terms field: a user types in vocabulary they expect to say, such as drug names, SKUs, or their own product names, and those words come back spelled as entered rather than transcribed phonetically. AssemblyAI reports that on its Universal-3.5 Pro model, typical short clips return in under a second, across 19 supported languages. Those figures describe AssemblyAI's own demo and reporting, not an outside of accuracy across arbitrary vocabulary.
- benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
Checking sign-in…
Loading comments…