Skip to main content
Server path: /deepgram | Type: Application | PCID required: Yes Speech-to-text and text-to-speech

Tools

ToolDescription
deepgram_transcribe_audioTranscribe audio file using Deepgram speech recognition
deepgram_transcribe_liveStart live audio transcription stream
deepgram_analyze_audioAnalyze audio for insights beyond transcription
deepgram_get_usageGet usage statistics for Deepgram API
deepgram_get_modelsGet available Deepgram models and languages
deepgram_get_projectsGet list of Deepgram projects

deepgram_transcribe_audio

Transcribe audio file using Deepgram speech recognition Parameters:
ParameterTypeRequiredDefaultDescription
audioUrlstringYesURL of audio file to transcribe
modelstringNo"nova-2"Deepgram model to use
languagestringNo"en-US"Language code
punctuatebooleanNotrueAdd punctuation to transcript
diarizebooleanNofalseEnable speaker diarization
numeralsbooleanNotrueConvert numbers to numerals
profanityFilterbooleanNofalseFilter profanity
redactstring[]NoRedaction options. Common: pci (credit card info), pii (personally identifiable info), phi (protected health info), numbers (numerical entities), ssn (social security numbers). For pre-recorded: supports all options plus specific entity types. For streaming: pci, phi, numbers, aggressive_numbers, ssn.
summarizebooleanNofalseGenerate summary
detectTopicsbooleanNofalseDetect topics
detectLanguagebooleanNofalseAuto-detect language

deepgram_transcribe_live

Start live audio transcription stream Parameters:
ParameterTypeRequiredDefaultDescription
modelstringNo"nova-2"Deepgram model to use
languagestringNo"en-US"Language code
punctuatebooleanNotrueAdd punctuation to transcript
interimResultsbooleanNotrueReturn interim results
endpointingnumberNoEndpointing sensitivity (0-500ms)
vadEventsbooleanNofalseVoice activity detection events
redactstring[]NoRedaction options for streaming. Supported: pci (credit card info), phi (protected health info), numbers (3+ consecutive numerals), aggressive_numbers (all numerals), ssn (social security numbers).

deepgram_analyze_audio

Analyze audio for insights beyond transcription Parameters:
ParameterTypeRequiredDefaultDescription
audioUrlstringYesURL of audio file to analyze
featuresstring[]YesAnalysis features to include
languagestringNo"en-US"Language code

deepgram_get_usage

Get usage statistics for Deepgram API Parameters:
ParameterTypeRequiredDefaultDescription
startDatestringYesStart date (YYYY-MM-DD format)
endDatestringYesEnd date (YYYY-MM-DD format)
includeRequestsbooleanNofalseInclude detailed request information

deepgram_get_models

Get available Deepgram models and languages Parameters:
ParameterTypeRequiredDefaultDescription
includeOutdatedbooleanNofalseInclude outdated models

deepgram_get_projects

Get list of Deepgram projects