Speech to text

deepgram-nova-3

File speech-to-text at Deepgram Pay As You Go. $0.0043 per minute for one language, $0.0052 per minute when language is multi. File diarization is included.

The call is Pascal’s. Send Authorization: Bearer with a Pascal key, and Pascal’s body: model, the file or the text, and params. Inside params, a Pascal name and the provider’s own name are both accepted. If both are sent and they disagree, the request is rejected. The provider’s SDK and host are not this API. The API request fee is not charged. A studio call needs a Creator plan or an API balance.

Native waveform: Deepgram field names and the shared names are both accepted.. Live price and availability are on GET /v1/models/deepgram-nova-3.

  • Deepgram names and the shared names are both valid. diarize and speaker_diarization are the same switch. keyterm and custom_terms are the same list. If both are sent and they disagree, the request is rejected.
  • The result keeps Deepgram’s utterances, words, paragraphs, summary, topics, intents, sentiments, entities, and search results, and also fills text and segments.
  • Summarization, sentiment, topics, and intents are billed from the token counts Deepgram returns: $0.30 per million input tokens and $0.60 per million output tokens.
  • Options Deepgram publishes on file transcription are accepted, including profanity_filter, numerals, dictation, measurements, filler_words, multichannel, detect_language, replace, search, keywords, callback, and mip_opt_out.

Parameters

FieldTypeDescription
params.language
Default en
stringBCP-47 tag, or multi. auto leaves Deepgram on English.
params.diarizebooleanSpeaker labels. speaker_diarization is the same switch. Included in the file price.
params.diarize_modelstringlatest, v1, or v2. Setting this turns diarization on.
params.modelstringnova-3, nova-3-general, nova-3-medical, or nova-3-pharma.
params.keytermstring or arrayKeyterm prompting. custom_terms is the same list. Adds $0.0013 per minute.
params.punctuatebooleanPunctuation and capitalization.
params.smart_formatbooleanIncluded in the file price.
params.utterancesbooleanSemantic segments with start, end, and speaker.
params.paragraphsbooleanParagraph breaks.
params.redactboolean or arrayRedaction. Adds $0.0020 per minute.
params.detect_entitiesbooleanEntity detection. Adds $0.0017 per minute.

Example

curl https://api.pascalai.co/v1/speech-to-text \
  -H "Authorization: Bearer $PASCAL_API_KEY" \
  -H "Idempotency-Key: $(uuidgen)" \
  -F model=deepgram-nova-3 \
  -F params='{"language":"en","diarize":true,"punctuate":true,"smart_format":true,"utterances":true,"keyterm":["Pascal"]}' \
  -F file=@input.wav

curl https://api.pascalai.co/v1/speech-to-text/$JOB_ID \
  -H "Authorization: Bearer $PASCAL_API_KEY"