Feature Overview

📘
To learn how to get up and running with Streaming Speech-to-Text, read the Streaming Speech-to-Text getting started guide.

Model Selection

Feature	Language(s)
Model	All available
Language	All available
Version	All available

Formatting

Feature	Language(s)
Smart Formatting	All available
Diarization	All available
Filler words	English (all available regions)
Numerals	English (all available regions)
Punctuation	All available
Profanity Filter	English (all available regions)
Redaction	English (all available regions)

Custom Vocabulary

Feature	Language(s)
Find and Replace	All available
Keywords	All available
Search	All available

Media Input Settings

Feature	Language(s)
Multichannel	All available
Sample rate	All available
Channels	All available

Results Processing

Feature	Language(s)
Callback	All available
Endpointing	All available
Utterance End	All available
Speech Started	All available
Interim results	All available
Tagging	All available
Extra Metadata	All available

Control Messages

Feature
Close Stream
Finalize
Keep Alive

Rate Limits

📘
For information on Deepgram's Concurrency Rate Limits, refer to our API Rate Limits Documentation.

Deepgram Self-Hosted

📘
Having challenges with performance and latency? Check out Deepgram's Self-Hosted Solution to get the benefits of running your own hosted instance of Deepgram.