How much does Deepgram Speech-to-Text cost per hour?

Deepgram’s Pay-As-You-Go pricing for our standard Nova model is approximately $0.46 per hour ($0.0077/min). For high-volume users on our Growth plan, the cost drops to roughly $0.39 per hour . This is significantly lower than major cloud providers, which often charge upwards of $1.44 per hour.

Does Deepgram charge for silence or round up audio time?

No. Deepgram uses true per-second billing . If your audio file is 14 seconds long, you pay for exactly 14 seconds. Many competitors round up to the nearest 15 seconds or full minute, which can inflate your actual invoice by 15-20%.

What is included in the $200 free credit?

Every new account receives $200 in free credit , which is equivalent to approximately 43,000 minutes (over 700 hours) of transcription using our Nova model. Unlike "free tiers" that expire after 12 months, this credit is available until you use it up, allowing you to prototype without time pressure.

How do you calculate costs for multichannel audio?

We bill based on the total duration of processed audio . If you process a 10-minute file with 2 channels (stereo), you are billed for 20 minutes of processing. This ensures you only pay for the exact amount of speech data analyzed.

What is the difference between Pay-As-You-Go and Growth plans?

Pay-As-You-Go requires no upfront commitment and bills monthly based on usage. The Growth plan requires a commitment (starting at $4k/year) but unlocks a ~20% discount on per-minute rates, higher concurrency limits, and priority support.

Are there extra fees for real-time streaming vs. pre-recorded audio?

No. Uniquely in the industry, Deepgram charges the same low rate for both real-time streaming and pre-recorded (batch) transcription. We do not charge a premium for the low-latency infrastructure required for streaming.

How does pricing work for the Voice Agent API?

The Voice Agent API is billed based on the duration of the conversation . It includes the orchestration of Speech-to-Text (listening), LLM processing (thinking), and Text-to-Speech (speaking). If you "Bring Your Own" LLM (e.g., GPT-4), you only pay Deepgram for the audio components and orchestration, often resulting in lower costs.

Do Audio Intelligence features cost extra?

Yes. Features like Summarization, Topic Detection, and Sentiment Analysis are billed per input unit (per hour or per 1,000 tokens) in addition to the transcription cost. However, because these run on our specialized small language models, they are highly cost-effective compared to running raw text through a generic LLM.

Is Text-to-Speech (Aura) billed by the minute or character?

Deepgram Aura TTS is billed by input character (e.g., $0.015 per 1,000 characters). This allows for precise cost control, as you are only charged for the text you actually convert to speech, regardless of how slowly or quickly the voice speaks.

Can I deploy Deepgram on-premise or in a private cloud?

Yes. For Enterprise customers with strict data sovereignty or security requirements (such as banking or healthcare), we offer self-hosted containers that you can deploy in your own VPC or on-premise hardware (requires NVIDIA GPUs).

What happens if I exceed my concurrency limit?

On the Pay-As-You-Go plan, requests exceeding your concurrency limit (e.g., 100 concurrent streams) may be queued or rejected. To avoid this, you can upgrade to a Growth or Enterprise plan, which offers custom concurrency limits to match your peak traffic needs.

Is Deepgram HIPAA and SOC 2 compliant?

Yes. Deepgram is SOC 2 Type 2 certified and HIPAA compliant . We can sign a Business Associate Agreement (BAA) for Enterprise customers handling sensitive healthcare data.

Do you offer volume discounts for high-usage applications?

Yes. If your application processes millions of minutes per month, our Enterprise plan offers significant volume discounts, custom model training, and dedicated account management.

Deepgram Pricing

Straightforward costs for projects at any scale.

Find the right plan with our clear, transparent, and flexible pricing. Whether you're just exploring or deploying at enterprise scale, our plans are built for innovation, not contract negotiation.

Speech to Text

Power your apps with world-class speech recognition in 30+ languages. Includes: Speaker Diarization, Smart formatting, Automatic Language Detection, Deep Search, Keyword Boosting, Multichannel Support, Redaction (PCI, PII) and Callbacks.

For detailed model, language, and feature availability, please refer to our Developer Documentation.

Model	Pay As You Go	Growth
Nova i Our highest performing model. Recommended for most use cases, especially audio with multiple languages, background noise, crosstalk and far field audio.	$0.0043/min	$0.0036/min
Enhanced i Recommended for lower word error rates than Base, high accuracy timestamps, and use cases that require keyword boosting.	$0.0145/min	$0.0115/min
Base i Recommended for large transcription volumes and high accuracy timestamps.	$0.0125/min	$0.0095/min
Whisper Large	$0.0048/min	$0.0048/min
Custom

Model	Pay As You Go
Nova i Our highest performing model. Recommended for most use cases, especially audio with multiple languages, background noise, crosstalk and far field audio.	$0.0043/min
Enhanced i Recommended for lower word error rates than Base, high accuracy timestamps, and use cases that require keyword boosting.	$0.0145/min
Base i Recommended for large transcription volumes and high accuracy timestamps.	$0.0125/min
Whisper Large	$0.0048/min
Custom

Rates listed above opt in to the Model Improvement Program

Text to Speech

Responsive, natural sounding text-to-speech to power your high throughput voicebots and conversational AI applications. Billed per character.

Model	Pay As You Go	Growth	Enterprise
Aura	$0.0150/1k characters	$0.0135/1k characters	Contact Sales

Model	Pay As You Go
Aura	$0.0150/1k characters

Rates listed above opt in to the Model Improvement Program

Audio Intelligence

Powered by task-specific language models. Can be used alongside or independently from Transcription. Accepts text or audio.

Model	Pay As You Go	Growth	Enterprise
Summarization	$0.0040	$0.0032	Contact Sales
Topic Detection	Get Early Access	Get Early Access
Sentiment Analysis
Intent Recognition

Sentiment Analysis
Model	Pay As You Go
Summarization	$0.0040
Topic Detection	Get Early Access
Intent Recognition

Rates listed above opt in to the Model Improvement Program.

Deepgram Pricing

Speech to Text

Text to Speech

Audio Intelligence

Frequently Asked Questions