Example output
{
  "metadata": {
    "duration": 25.9
  },
  "results": {
    "summary": {
      "short": "Speaker discusses significance of first all-female spacewalk..."
    },
    "channels": [
      {
        "alternatives": [
          {
            "transcript": "Yeah...",
            "entities": [
              {
                "label": "ORDINAL",
                "value": "first"
              }
            ]
          }
        ]
      }
    ]
  }
}
inference · 3.7s

Deepgram Nova-3 · Speech to Text Pro

Object·nova-3·by Deepgram

Deepgram Nova-3 Pro delivers advanced speech-to-text with built-in AI features like summarization, topic and entity detection, sentiment and intent analysis, plus smart formatting and redaction.

Runtime (p50)
-
Estimated price
From $0.000073
Call the API
prediction.sh
sh
curl -X POST \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Content-Type: application/json" \
  --data '{
    "model": "deepgram-nova-3-speech-to-text-pro",
    "version": "0.0.1",
    "input": {
        "model": "nova-3",
        "redact": "pci",
        "topics": true,
        "diarize": true,
        "intents": true,
        "media_url": "https://storage.googleapis.com/magicpoint/inputs/deepgram-nova-3-stt-pro-input.wav",
        "punctuate": true,
        "sentiment": true,
        "summarize": true,
        "utt_split": 0.8,
        "paragraphs": true,
        "utterances": true,
        "filler_words": true,
        "smart_format": true,
        "language_code": "auto",
        "detect_entities": true,
        "custom_topic_mode": "extended",
        "custom_intent_mode": "extended"
    },
    "webhook_url": ""
}' \
  https://api.eachlabs.ai/v1/prediction/

Related models

4 models
* FAQ

About Deepgram Nova-3 · Speech to Text Pro

01 / 03

What is Deepgram Nova-3 Speech to Text Pro?

Deepgram Nova-3 Speech to Text Pro is an advanced ASR model by Deepgram offering best-in-class transcription accuracy, low latency, and support for multiple languages. The Pro variant includes enhanced punctuation, speaker confidence scores, and optimized noise robustness.