# Deepgram Review

> Deepgram review: Developer-first realtime STT API for voice agents and transcription at scale. Verified pricing, features, and the strongest alternatives in.

![Deepgram logo](https://www.versusref.com/logos/deepgram.png)

Developer-first realtime STT API for voice agents and transcription at scale

Among the 43 speech-to-text tools we track, Deepgram has the 26th-widest language coverage.

See pricing · facts verified Jul 20, 2026

## What we know about Deepgram

Deepgram is a speech-to-text apis platform: developer-first realtime STT API for voice agents and transcription at scale. This profile tracks every Deepgram fact we have verified, each linked to a primary source and dated.

Deepgram does not publish a public per-minute rate we have been able to verify, so the price figure here stays blank until we can confirm one. We would rather show nothing than a guessed number; if you have a current Deepgram quote, it is the fastest way for us to close that gap.

On capabilities, Deepgram covers language auto-detection, word-level timestamps, custom vocabulary / keyterm boosting, entity detection, sentiment analysis, and summarization endpoint. Each of those is verified against Deepgram's own docs or dashboard, not marketing copy.

For compliance, with Deepgram: SOC 2 Type II is in place. If you are in a regulated space, confirm the current posture with Deepgram before you commit, since these change plan by plan.

Placed against the 43 speech-to-text tools we track, Deepgram's strongest showing is the 26th-widest language coverage - a spread worth weighing against your own priorities.

In total we track 29 verified facts for Deepgram today, each linking the primary source it came from so you can check our work - and vendor claims we have not measured ourselves are labeled as such on the Deepgram fact sheet below.

## Deepgram pricing

Published rates: batch $0.0043/min · streaming $0.0048/min, verified Jul 20, 2026 ([source](https://deepgram.com/pricing)).

| Monthly volume | Batch bill | Streaming bill |
| --- | --- | --- |
| 1K min/mo | $4.30 | $4.80 |
| 10K min/mo | $43 | $48 |
| 100K min/mo | $430 | $480 |

Sticker rates only; diarization and PII redaction add-ons price in the stack builder.

## Fact sheet

### Pricing

| Fact | Value | Verified | Source |
| --- | --- | --- | --- |
| Batch price per audio minute | 0.004 $/audio-min | Jul 20 | [source](https://deepgram.com/pricing) |
| Streaming price per audio minute | 0.005 $/audio-min | Jul 20 | [source](https://deepgram.com/pricing) |
| Pricing model | usage | Jul 20 | [source](https://deepgram.com/pricing) |
| Free tier quota | $200 credit, no expiration, no credit card required | Jul 20 | [source](https://deepgram.com/pricing) |
| Published volume discounts | Growth: save up to 20% with $4K+/yr prepaid credits | Jul 20 | [source](https://deepgram.com/pricing) |
| Concurrency on base plan | PAYG STT: 50 REST, 150 websocket concurrent | Jul 20 | [source](https://developers.deepgram.com/reference/api-rate-limits) |

### Capabilities

| Fact | Value | Verified | Source |
| --- | --- | --- | --- |
| WER (third-party benchmark) | 5.18 % WER | Jul 20 | [source](https://artificialanalysis.ai/speech-to-text) |
| WER (vendor-claimed) | ~5.26 % WER | Jul 20 | [source](https://deepgram.com/learn/introducing-nova-3-speech-to-text-api) |
| Streaming latency (vendor-claimed) | ~300 ms | Jul 20 | [source](https://deepgram.com/product/speech-to-text) |
| Languages supported | ~50 | Jul 20 | [source](https://deepgram.com/product/speech-to-text) |
| Language auto-detection | ✓  Yes | Jul 20 | [source](https://developers.deepgram.com/docs/language-detection) |
| Speaker diarization | ◑  Paid add-on | Jul 20 | [source](https://deepgram.com/pricing) |
| Diarization add-on price per minute | 0.002 $/audio-min | Jul 20 | [source](https://deepgram.com/pricing) |
| PII redaction | ◑  Paid add-on | Jul 20 | [source](https://deepgram.com/pricing) |
| PII redaction add-on price per minute | 0.002 $/audio-min | Jul 20 | [source](https://deepgram.com/pricing) |
| Word-level timestamps | ✓  Yes | Jul 20 | [source](https://developers.deepgram.com/docs/utterances) |
| Custom vocabulary / keyterm boosting | ✓  Yes | Jul 20 | [source](https://developers.deepgram.com/docs/keyterm) |
| Entity detection | ✓  Yes | Jul 20 | [source](https://deepgram.com/pricing) |
| Sentiment analysis | ✓  Yes | Jul 20 | [source](https://deepgram.com/pricing) |
| Summarization endpoint | ✓  Yes | Jul 20 | [source](https://developers.deepgram.com/docs/summarization) |
| Priced audio-intelligence add-ons | Summarization, topics, sentiment, intents; token-priced | Jul 20 | [source](https://deepgram.com/pricing) |

### Compliance & trust

| Fact | Value | Verified | Source |
| --- | --- | --- | --- |
| HIPAA BAA available | ✓  Yes | Jul 20 | [source](https://developers.deepgram.com/trust-security/data-privacy-compliance) |
| SOC 2 Type II | ✓  Yes | Jul 20 | [source](https://developers.deepgram.com/trust-security/data-privacy-compliance) |
| GDPR / EU data residency | ✓  Yes | Jul 20 | [source](https://developers.deepgram.com/trust-security/data-privacy-compliance) |
| Self-host / on-prem option | ✓  Yes | Jul 20 | [source](https://developers.deepgram.com/docs/self-hosted-introduction) |

### Build experience

| Fact | Value | Verified | Source |
| --- | --- | --- | --- |
| Official SDKs | JS/TS, Python, .NET, Go, Java, Rust | Jul 20 | [source](https://developers.deepgram.com/sdks) |
| Max file size / duration | 2 GB max file; 10 min processing timeout (Nova) | Jul 20 | [source](https://developers.deepgram.com/docs/pre-recorded-audio) |
| Supported audio formats | 100+ formats incl MP3, MP4, WAV, FLAC, Ogg, Opus, WebM | Jul 20 | [source](https://developers.deepgram.com/docs/supported-audio-formats) |
| Websocket streaming API | ✓  Yes | Jul 20 | [source](https://developers.deepgram.com/reference/api-rate-limits) |

## Deepgram head-to-head

| Comparison | Record | Per use case |
| --- | --- | --- |
| [Deepgram vs AssemblyAI](https://www.versusref.com/stt/assemblyai-vs-deepgram/) | won 0 · lost 3 · tied 3 | Call Centers: lost; Developers: lost; Dictation: tie; Medical: tie; Meetings: lost; Self-Hosted: tie |
| [Deepgram vs OpenAI Whisper (API)](https://www.versusref.com/stt/deepgram-vs-whisper/) | won 5 · lost 1 · tied 1 | Call Centers: won; Developers: won; Dictation: tie; Medical: won; Meetings: won; Self-Hosted: lost; Voice Agents: won |
| [Deepgram vs ElevenLabs Scribe](https://www.versusref.com/stt/deepgram-vs-elevenlabs-scribe/) | won 2 · lost 0 · tied 1 | Developers: won; Dictation: tie; Self-Hosted: won |
| [Deepgram vs Mistral Voxtral Transcribe](https://www.versusref.com/stt/deepgram-vs-voxtral/) | won 4 · lost 2 · tied 1 | Call Centers: won; Developers: won; Dictation: tie; Medical: won; Meetings: won; Self-Hosted: lost; Voice Agents: lost |
| [Deepgram vs Speechmatics](https://www.versusref.com/stt/deepgram-vs-speechmatics/) | won 2 · lost 1 · tied 2 | Dictation: tie; Medical: won; Meetings: lost; Self-Hosted: tie; Voice Agents: won |
| [Deepgram vs Gladia](https://www.versusref.com/stt/deepgram-vs-gladia/) | won 4 · lost 1 · tied 2 | Call Centers: won; Developers: won; Dictation: tie; Medical: won; Meetings: lost; Self-Hosted: tie; Voice Agents: won |
| [Deepgram vs Soniox](https://www.versusref.com/stt/deepgram-vs-soniox/) | won 3 · lost 3 · tied 1 | Call Centers: lost; Developers: lost; Dictation: tie; Medical: won; Meetings: won; Self-Hosted: won; Voice Agents: lost |
| [Deepgram vs Amazon Transcribe](https://www.versusref.com/stt/amazon-transcribe-vs-deepgram/) | won 3 · lost 0 · tied 1 | Developers: won; Dictation: tie; Self-Hosted: won; Voice Agents: won |
| [Deepgram vs Azure AI Speech (STT)](https://www.versusref.com/stt/azure-speech-vs-deepgram/) | won 2 · lost 1 · tied 3 | Developers: lost; Dictation: tie; Medical: won; Meetings: tie; Self-Hosted: tie; Voice Agents: won |
| [Deepgram vs Cartesia Ink](https://www.versusref.com/stt/cartesia-ink-vs-deepgram/) | won 4 · lost 1 · tied 2 | Call Centers: won; Developers: won; Dictation: tie; Medical: won; Meetings: won; Self-Hosted: tie; Voice Agents: lost |
| [Deepgram vs Google Cloud Speech-to-Text](https://www.versusref.com/stt/deepgram-vs-google-stt/) | won 5 · lost 0 · tied 2 | Call Centers: won; Developers: won; Dictation: tie; Medical: won; Meetings: won; Self-Hosted: tie; Voice Agents: won |
| [Deepgram vs Rev AI](https://www.versusref.com/stt/deepgram-vs-rev/) | won 1 · lost 1 · tied 1 | Developers: lost; Dictation: tie; Voice Agents: won |
| [Deepgram vs NVIDIA Parakeet / Riva](https://www.versusref.com/stt/deepgram-vs-nvidia-parakeet/) | won 3 · lost 1 · tied 1 | Developers: won; Dictation: tie; Medical: won; Self-Hosted: lost; Voice Agents: won |
| [Deepgram vs Cohere Transcribe](https://www.versusref.com/stt/cohere-transcribe-vs-deepgram/) | won 5 · lost 1 · tied 1 | Call Centers: won; Developers: won; Dictation: tie; Medical: won; Meetings: won; Self-Hosted: lost; Voice Agents: won |
| [Deepgram vs xAI Grok Speech-to-Text](https://www.versusref.com/stt/deepgram-vs-grok-stt/) | won 3 · lost 0 · tied 1 | Dictation: tie; Medical: won; Self-Hosted: won; Voice Agents: won |
| [Deepgram vs Smallest.ai Pulse](https://www.versusref.com/stt/deepgram-vs-smallest-pulse/) | won 2 · lost 2 · tied 2 | Call Centers: lost; Developers: won; Dictation: tie; Medical: won; Meetings: lost; Self-Hosted: tie |
| [Deepgram vs IBM watsonx Speech to Text](https://www.versusref.com/stt/deepgram-vs-ibm-watson-stt/) | won 2 · lost 0 · tied 2 | Developers: won; Dictation: tie; Self-Hosted: tie; Voice Agents: won |

Source: https://www.versusref.com/stt/tools/deepgram/