# Deepgram | Nova-3 | Speech to Text

Deepgram Nova-3 accurately transcribes pre-recorded audio with word-level timestamps, speaker diarization, and automatic language detection.

## API Information

- **Model Slug:** deepgram-nova-3-speech-to-text
- **Branded URL:** https://www.eachlabs.ai/deepgram/nova-3/deepgram-nova-3-speech-to-text
- **Provider:** Deepgram
- **Category:** Voice to Text
- **Output Type:** object
- **Status:** active
- **Base Cost:** $0.0000725–$0.000086666667 per second, depending on language code
- **Estimated Processing Time:** 10 seconds
- **Interactive Demo:** https://www.eachlabs.ai/ai-models/deepgram-nova-3-speech-to-text

## Pricing

- **Charge Type:** dynamic
- **Estimate:** $0.0000725–$0.000086666667 per second, depending on language code
- **Pricing Details:** multi: $0.000087
- **Pricing Details:** default: $0.000073

## 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. The file must be publicly accessible or an EachLabs-uploaded file. |
| 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 code). auto: automatic detection (recommended). multi: multilingual audio with up to 10 languages. tr: Turkish, ur: Urdu, en: English, nl: Dutch, uk: Ukrainian, es: Spanish, ar: Arabic, de: German, fr: French, it: Italian, ja: Japanese, ko: Korean, pt: Portuguese, ru: Russian, zh: Chinese, hi: Hindi. Nova-3 supports 47+ languages total. |
| diarize | boolean | No | true | - | Identify different speakers in the audio. When enabled, each word in the response includes a speaker ID (integer) and speaker_confidence score. Essential for multi-speaker audio like meetings, interviews, and phone calls. |
| multichannel | boolean | No | false | - | Transcribe each audio channel independently. Enable when each channel contains a single speaker (e.g., stereo call recordings with one speaker per channel). Max 5 channels. Each word includes a channel index. |
| smart_format | boolean | No | true | - | Auto-format currency amounts, phone numbers, email addresses, dates, and other entities for enhanced readability. Recommended for most use cases. |
| punctuate | boolean | No | true | - | Add punctuation marks and capitalization to the transcript. Produces more readable output. Recommended for most use cases. |
| model | string | No | nova-3 | nova-3, nova-2 | Deepgram speech recognition model. nova-3: latest generation, best accuracy, 47+ languages, recommended for all use cases. nova-2: previous generation, still available for backward compatibility. |

## Example Request

```bash
curl -X POST https://api.eachlabs.ai/v1/prediction/ \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "deepgram-nova-3-speech-to-text",
  "input": {
    "media_url": "https://cdn-us.eachlabs.ai/defaults/61ac7afa807c63957303911d27c8ca2ba57f6ccee1683fb1bc57006c50c8ca6e.mp3"
  }
}'
```

## 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 "Authorization: Bearer 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 `Authorization` bearer token |
| `404` | Unknown model or prediction id |
| `429` | Rate limit — 100 creates / min, 10 concurrent per key |
| `5xx` | Retry with backoff |

## Overview

**Deepgram | Nova-3 | Speech to Text Overview**

Deepgram | Nova-3 | Speech to Text is a high-performance speech recognition API that converts audio into accurate text transcriptions with word-level timestamps and speaker identification. Built by Deepgram, Nova-3 represents the latest generation of their speech-to-text technology, designed specifically for real-time and streaming applications where latency matters. The model's primary differentiator is its exceptional speed: Nova-3 achieves 441.6x real-time processing with sub-300ms streaming latency, making it the fastest commercial speech-to-text solution available. This combination of accuracy and speed makes Deepgram | Nova-3 | Speech to Text ideal for voice agents, IVR systems, and any application requiring conversational responsiveness.

## Usage Notes

- API Base URL: `https://api.eachlabs.ai/v1`
- Authentication: send `Authorization: Bearer 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.