Deepgram Post Listen Transcribe

From 5.16 credits/min
ae_deepgram_post_listen_transcribe

Submit prerecorded audio or video to Deepgram for background transcription. Pass the file URL as request_body.url. The submission returns an AE task_id. Use ae_get_result with that ID to retrieve the complete transcript, timestamps and metadata. Set diarize_model=v2 to include speaker labels. language=multi enables multilingual transcription. Billing uses actual audio duration.

TranscriptAudio
From 5.16 credits/min

Pricing

From 5.16 credits/min
ModePer minutePer second
Monolingual5.16 credits0.086 credits
Multilingual6.24 credits0.104 credits

Billed per second of actual audio. Speaker diarization is included.

Parameters

Optional parameters use the platform defaults when omitted.

ParameterTypeRequiredDescriptionDefault
tagstring | arrayNoLabel your requests for the purpose of identification during usage reporting—
extrastring | arrayNoArbitrary key-value pairs that are attached to the API response for usage in downstream processing—
modelstringNoAI model used to process submitted audiobase-general
redactstring | arrayNoRedaction removes sensitive information from your transcripts—
searchstring | arrayNoSearch for terms or phrases in submitted audio—
topicsbooleanNoDetect topics throughout a transcript or textfalse
intentsbooleanNoRecognizes speaker intent throughout a transcript or textfalse
keytermarrayNoKey term prompting can boost or suppress specialized terminology and brands. Only compatible with Nova-3—
replacestring | arrayNoSearch for terms or phrases in submitted audio and replaces them—
versionstringNoVersion of an AI model to uselatest
encodingenum (linear16 | flac | mulaw | amr-nb | amr-wb | opus | speex | g729)NoSpecify the expected encoding of your submitted audio—
keywordsstring | arrayNoKeywords can boost or suppress specialized terminology and brands—
languagestringNoThe [BCP-47 language tag](https://tools.ietf.org/html/bcp47) that hints at the primary spoken language. Depending on the Model and API endpoint you choose only certain languages are availableen
numeralsbooleanNoNumerals converts numbers from written format to numerical formatfalse
dictationbooleanNoDictation mode for controlling formatting with dictated speechfalse
punctuatebooleanNoAdd punctuation and capitalization to the transcriptfalse
sentimentbooleanNoRecognizes the sentiment throughout a transcript or textfalse
summarizestring | booleanNoSummarize content. For Listen API, supports string version option. For Read API, accepts boolean only.—
utt_splitnumberNoSeconds to wait before detecting a pause between words in submitted audio0.8
paragraphsbooleanNoSplits audio into paragraphs to improve transcript readabilityfalse
utterancesbooleanNoSegments speech into meaningful semantic unitsfalse
mip_opt_outbooleanNoOpts out requests from the Deepgram Model Improvement Program. Refer to our Docs for pricing impacts before setting this to true. https://dpgr.am/deepgram-mipfalse
custom_topicstring | arrayNoCustom topics you want the model to detect within your input audio or text if present Submit up to `100`.—
filler_wordsbooleanNoFiller Words can help transcribe interruptions in your audio, like "uh" and "um"false
measurementsbooleanNoSpoken measurements will be converted to their corresponding abbreviationsfalse
multichannelbooleanNoTranscribe each audio channel independentlyfalse
request_bodyobjectNoAudio file URL to transcribe—
request_body.urlstringYesNo parameter description.—
smart_formatbooleanNoApply formatting to transcript output. When set to true, additional formatting will be applied to transcripts to improve readabilityfalse
custom_intentstring | arrayNoCustom intents you want the model to detect within your input audio if present—
diarize_modelenum (latest | v1 | v2)NoEnable speaker attribution using a model version: latest, v1, or v2. For prerecorded audio, use v2 to pin the newer model. Omit this parameter when speaker labels are unnecessary.—
detect_entitiesbooleanNoIdentifies and extracts key entities from content in submitted audiofalse
detect_languageboolean | arrayNoIdentifies the dominant language spoken in submitted audio—
profanity_filterbooleanNoProfanity Filter looks for recognized profanity and converts it to the nearest recognized non-profane word or removes it from the transcript completelyfalse
custom_topic_modeunknownNoSets how the model will interpret strings submitted to the `custom_topic` param. When `strict`, the model will only return topics submitted using the `custom_topic` param. When `extended`, the model will return its own detected topics in addition to those submitted using the `custom_topic` paramextended
custom_intent_modeunknownNoSets how the model will interpret intents submitted to the `custom_intent` param. When `strict`, the model will only return intents submitted using the `custom_intent` param. When `extended`, the model will return its own detected intents in the `custom_intent` param.extended

Invocation example

Parameter template

Generated from the parameter configuration. Replace the sample values before use.

POST/agent-api/v1/tool/executeX-Api-Key
Request coderequest.sh
curl --request POST \
  --url 'https://agentearth.ai/agent-api/v1/tool/execute?tool_name=ae_deepgram_post_listen_transcribe' \
  --header "X-Api-Key: $AGENTEARTH_API_KEY" \
  --header 'Content-Type: application/json' \
  --data-raw '{
  "params": {}
}'
Response exampleresponse.json
JSON
"validating \"arguments\": validating root: required: missing properties: [\"url\"]"

Guides using this tool

See how this tool fits into real workflows