Features
| Feature | Pre-recorded? | Streaming? | Language(s) | ||
|---|---|---|---|---|---|
| Tier | Yes | Yes | Depends on Model | ||
| Model | |||||
general |
Enhanced | Yes | Yes | en, en-US, es |
|
general |
Base | Yes | Yes | All available | |
phonecall |
Enhanced beta | Yes | Yes | en, en-US |
|
phonecall |
Base | Yes | Yes |
en, en-US, en-IN,
en-GB
|
|
voicemail |
Base | Yes | Yes | en, en-US |
|
meeting |
Enhanced beta | Yes | Yes | en, en-US |
|
meeting |
Base | Yes | Yes | en, en-US |
|
finance |
Enhanced | Yes | Yes | en, en-US |
|
finance |
Base | Yes | Yes | en, en-US |
|
conversationalai |
Base | Yes | Yes | en, en-US |
|
video |
Base | Yes | Yes | en, en-US |
|
| Custom models | Enhanced & Base | Yes | Yes | Available in the language they are built in. | |
| Version | Yes | Yes | All available | ||
| Language Detection | Yes | No | All not in beta | ||
| Punctuation | Yes | Yes | All available | ||
| Profanity Filter | Yes | Yes | English (all available regions) | ||
| Redaction | Yes | Yes | English (all available regions) | ||
| Diarization | Yes | Yes | All available | ||
| Smart Format | Yes | Yes | English (all available regions) | ||
| Multichannel | Yes | Yes | All available | ||
| Alternatives | Yes | Yes | English (all available regions) | ||
| Numerals | Yes | Yes | English (all available regions) | ||
| Search | Yes | Yes | All available | ||
| Find and Replace | Yes | Yes | All available | ||
| Callback | Yes | Yes | All available | ||
| Keywords | Yes | Yes | All available | ||
| Paragraphs | Yes | No | All in which words are delimited by spaces | ||
| Summarization | Yes | No | English (all available regions) | ||
| Topic Detection | Yes | No | English (all available regions) | ||
| Utterances | Yes | No | All available | ||
| Interim Results | No | Yes | All available | ||
| Endpointing | No | Yes | All available | ||
| Encoding | No | Yes | All available | ||
| Channels | No | Yes | All available | ||
| Sample Rate | No | Yes | All available | ||
Guides
Tier
Learn about Deepgram's Tier feature, which allows you to associate your API requests with a specific tier.
Model
Learn about Deepgram's Model feature, which allows you to supply a model to use to process submitted audio.
Version
Learn about Deepgram's Sample Rate feature, which allows you to specify the version of the model you want to use to process your submitted audio.
Language
Learn about Deepgram's Language feature, which allows you to supply a BCP-47 language tag that hints at the primary spoken language of submitted audio.
Punctuation
Learn about Deepgrams' punctuation feature, which adds punctuation and capitalization to your transcript.
Language Detection
Learn about Deepgram's Language Detection feature, which identifies the dominant language spoken in submitted audio.
Profanity Filter
Learn about Deepgram's Profanity Filter feature, which looks for recognized profanity and converts it to the nearest recognized non-profane word or removes it from the transcript completely.
Redaction
Learn about Deepgram's Redaction feature, which redacts sensitive information, replacing redacted content with asterisks.
Diarization
Learn about Deepgram's Diarize feature, which recognizes speaker changes in submitted audio.
Multichannel
Learn about Deepgram's Multichannel feature, which transcribes each channel in submitted audio independently.
Smart Format
Learn about Deepgram's Smart Format feature, which formats transcripts to improve readability.
Numerals
Learn about Deepgram's numerals feature, which converts numbers from written format to numerical format.
Search
Learn about Deepgram's Search feature, which searches for terms or phrases in submitted audio.
Find and Replace
Learn about Deepgram's Find and Replace feature, which searches for terms or phrases in submitted audio and replaces them.
Callback
Learn about Deepgram's Callback feature, which allows you to have your submitted audio processed asynchronously.
Paragraphs
Learn about Deepgram's Paragraphs feature, which splits audio into paragraphs to improve transcript readability.
Keywords
Learn about Deepgram's Keyword feature, which allows you to boost or suppress Out-of-vocabulary (OOV) keywords in submitted audio.
Utterances
Learn about Deepgram's utterances feature, which segments speech into meaningful semantic units.
Utterance Split
Learn about Deepgram's Utterance Split feature, which detects pauses between words in submitted audio. Used when the Utterances feature is enabled for pre-recorded audio.
Interim Results
Learn about Deepgram's Interim Results feature, which provides preliminary results for streaming audio to solve the need for immediate results combined with high levels of accuracy.
Summarization
Learn about Deepgram's Summarization feature, which summarizes sections of content in submitted audio.
Topic Detection
Learn about Deepgram's Topic Detection feature, which identifies and extracts key topics from content in submitted audio.
Endpointing
Learn about Deepgram's Endpointing feature, which returns transcripts when pauses in speech are detected.
Encoding
Learn about Deepgram's Encoding feature, which allows you to specify the expected encoding of your submitted audio.
Channels
Learn about Deepgram's Channels feature, which allows you to specify the number of independent audio channels your submitted audio contains. Used when the Encoding feature is also being used to submit streaming raw audio.
Sample Rate
Learn about Deepgram's Sample Rate feature, which allows you to specify the sample rate of your submitted audio. Required when the Encoding feature is also being used to submit streaming raw audio.
Tagging
Learn about Deepgram's Tagging feature, which allows you to label your requests for the purpose of identification during usage reporting.