# Deepgram | Nova-3 | Speech to Text Pro

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.

## API Information

- **Model Slug:** deepgram-nova-3-speech-to-text-pro
- **Branded URL:** https://www.eachlabs.ai/deepgram/nova-3/deepgram-nova-3-speech-to-text-pro
- **Provider:** Deepgram
- **Category:** Voice to Text
- **Output Type:** object
- **Status:** active
- **Version:** 0.0.1
- **Base Cost:** Entity detection + Redaction — $0.00803/min
- **Estimated Processing Time:** 0 seconds
- **Last Updated:** 2026-04-02
- **Interactive Demo:** https://www.eachlabs.ai/ai-models/deepgram-nova-3-speech-to-text-pro

## Pricing

- **Charge Type:** dynamic
- **Pricing Details:** Entity detection + Redaction — $0.00803/min

### Pricing Rules

| Condition | Pricing |
| --- | --- |
| language_code eq_i "multi" AND detect_entities eq_i "true" AND redact in_i "pci,pii,phi,numbers" | Multilingual + Entity detection + Redaction — $0.00888/min |
| language_code eq_i "multi" AND detect_entities eq_i "true" | Multilingual + Entity detection — $0.00740/min |
| language_code eq_i "multi" AND redact in_i "pci,pii,phi,numbers" | Multilingual + Redaction — $0.00719/min |
| language_code eq_i "multi" | Multilingual (base) — $0.00520/min |
| detect_entities eq_i "true" AND redact in_i "pci,pii,phi,numbers" | Entity detection + Redaction — $0.00803/min |
| detect_entities eq_i "true" | Entity detection (includes Intel bundle) — $0.00655/min |
| redact in_i "pci,pii,phi,numbers" | Redaction only — $0.00634/min |
| Rule 8 | Base (mono) — $0.00435/min |

## Input Schema

| Parameter | Type | Required | Default | Constraints | Description |
|-----------|------|----------|---------|-------------|-------------|
| media_url | string | Yes | - | - | Audio file URL to transcribe. Supports mp3, wav, m4a, flac, ogg, webm, mp4, and 100+ audio formats. |
| language_code | string | No | auto | auto,multi,tr,ur,en,nl,uk,es,ar,de,fr,it,ja,ko,pt,ru,zh,hi,bn,cs,da,fi,el,he,hu,id,ms,no,pl,ro,sk,sv,ta,te,th,vi | Language of the audio (BCP-47). auto: automatic detection. multi: multilingual (up to 10 languages). 47+ languages supported. |
| diarize | boolean | No | true | - | Identify different speakers. Each word includes speaker ID and confidence score. |
| summarize | boolean | No | true | - | Generate a short text summary of the transcript content. |
| detect_entities | boolean | No | true | - | Extract named entities: persons, organizations, locations, dates, ordinals. Returns label, value, confidence per entity. |
| sentiment | boolean | No | true | - | Analyze sentiment per segment (positive/negative/neutral with score) and overall average. |
| topics | boolean | No | true | - | Detect topics discussed. Returns topic labels with confidence scores per segment. |
| intents | boolean | No | true | - | Detect speaker intents. Returns intent labels with confidence scores per segment. |
| utterances | boolean | No | true | - | Segment speech into semantic units. Returns array with start/end times, speaker, channel, transcript per utterance. |
| paragraphs | boolean | No | true | - | Split transcript into paragraphs with sentences. Requires punctuate=true. |
| smart_format | boolean | No | true | - | Auto-format currency, phone numbers, emails, dates for readability. |
| punctuate | boolean | No | true | - | Add punctuation and capitalization to transcript. |
| numerals | boolean | No | false | - | Convert spoken numbers to digits (e.g., three hundred to 300). |
| profanity_filter | boolean | No | false | - | Replace profanity with asterisks in transcript. |
| filler_words | boolean | No | true | - | Include filler words (uh, um) in transcript. Useful for conversation analysis. |
| redact | string | No | false | false,pci,pii,numbers | Redact sensitive info. false: none. pci: credit cards. pii: SSN/phone/email. numbers: all numeric sequences. |
| multichannel | boolean | No | false | - | Transcribe each audio channel independently. Max 5 channels. |
| utt_split | number | No | 0.8 | - | Seconds of silence to split utterances. Lower = more splits. Only when utterances=true. Default 0.8. |
| model | string | No | nova-3 | nova-3,nova-2 | Deepgram model. nova-3: latest, best accuracy. nova-2: previous generation. |

## Example Request

```bash
curl -X POST https://api.eachlabs.ai/v1/prediction/ \
  -H "X-API-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "deepgram-nova-3-speech-to-text-pro",
  "input": {
    "media_url": "https://storage.googleapis.com/magicpoint/inputs/deepgram-nova-3-stt-pro-input.wav"
  }
}'
```

## Output Schema

Response returned by `GET /v1/prediction/{id}` when the job completes:

```json
{
  "status": "success",
  "predictionID": "string",
  "output": "object",
  "metrics": {
    "predict_time": "number (seconds)"
  }
}
```

## Polling

```bash
curl https://api.eachlabs.ai/v1/prediction/{PREDICTION_ID} \
  -H "X-API-Key: YOUR_API_KEY"
```

| Status | Meaning |
|--------|---------|
| `processing` | Still running — poll again |
| `success` | Done — read `output` |
| `error` | Failed — read `message` / `details` |

## Webhook (alternative to polling)

Pass `"webhook_url": "https://your.host/path"` in the create request. Eachlabs POSTs this payload when the job ends:

```json
{
  "exec_id": "prediction-uuid",
  "status": "succeeded",
  "output": "https://...",
  "error": ""
}
```

`status` is `"succeeded"` or `"failed"`. `exec_id` equals the `predictionID` from create. Return 2xx within 30 seconds.

## Errors

Error body: `{ "status": "error", "message": "...", "details": "..." }`

| Code | Meaning |
|------|---------|
| `400` | Invalid input |
| `401` | Missing / invalid `X-API-Key` |
| `404` | Unknown model or prediction id |
| `429` | Rate limit — 100 creates / min, 10 concurrent per key |
| `5xx` | Retry with backoff |

## Usage Notes

- API Base URL: `https://api.eachlabs.ai/v1`
- Authentication: send `X-API-Key: YOUR_API_KEY`. Generate a key from the Eachlabs dashboard at https://www.eachlabs.ai/dashboard/api-keys.
- File-typed parameters (`*_url`, `image_url`, `video_url`, `audio_url`, etc.) accept publicly-reachable HTTPS URLs only. Upload your asset first (GCS / S3 / your CDN) and pass the resulting URL. Data-URIs and localhost URLs are rejected.
- For structured parameters (arrays / objects) send real JSON values, not stringified payloads.
- Monetary values are reported in USD; per-token / per-megapixel rates may be billed in micro-cents internally.
- Prefer `webhook_url` over polling for long-running predictions — see the Webhook Callback section.