Unbeatable Pricing. Powerful Speech-to-Text API.
Power your apps with world-class automatic speech recognition models. Effortlessly accurate. Blazing fast. Enterprise-ready scale. Hands-down the best price.
Pay As You Go
$200 free credit then pay-as-you-go. No minimums. No expiration. No credit card required.
Start Free
Best for:
Individual developers building new voice apps.
Models:
-
Deepgram Nova (Batch)Starting at $0.0044/min
-
Deepgram Nova (Streaming)Starting at $0.0059/min
-
Deepgram Whisper Cloud (Batch)Starting at $0.0048/min
Features:
-
Transcription (30+ languages)
-
Language detection
-
Speaker diarization
-
Word-level timestamps
-
Smart formatting
-
Audio waveform deep search
-
Rate limit (100 concurrent requests)
Add ons:
Pricing: $0.0043/min
Build audio understanding features with language AI models.
-
Summarization
-
Entity detection
-
PII redaction
-
Topic detection
Support:
-
Community support
Growth
Save ~20% with pre-paid credits for the year. No credit card required.
$350+/mo
Best for:
Teams starting to grow their voice apps.
Models:
-
Deepgram Nova (Batch)Starting at $0.0036/min
-
Deepgram Nova (Streaming)Starting at $0.0049/min
-
Deepgram Whisper Cloud (Batch)Starting at $0.0038/min
Features:
-
Transcription (30+ languages)
-
Language detection
-
Speaker diarization
-
Word-level timestamps
-
Smart formatting
-
Audio waveform deep search
-
Higher rate limits
-
Multi-channel support
-
Early access to new features
Add ons:
Pricing: $0.0035/min
Build audio understanding features with language AI models.
-
Summarization
-
Entity detection
-
PII redaction
-
Topic detection
Support:
-
Community support
Premium
Scale effortlessly with best pricing, reliability, and performance.
Let’s Talk
Best for:
Teams building voice-enabled products at scale.
Models:
-
Deepgram Nova (Batch)
-
Deepgram Nova (Streaming)
-
Deepgram Whisper Cloud
-
Deepgram Nova Lite
-
Access to Fine-Tuned Models
Features:
-
Transcription (30+ languages)
-
Language detection
-
Speaker diarization
-
Word-level timestamps
-
Smart formatting
-
Audio waveform deep search
-
Multi-channel support
-
Best pricing (volume discounts)
-
No rate limits
-
Priority access to new features
-
Premium integrations
-
On-prem or VPC deployment
-
Discount off-peak processing
-
Premium-grade security
Add ons:
Pricing: Volume Discount
Build audio understanding features with language AI models.
-
Summarization
-
Entity detection
-
PII redaction
-
Topic detection
Support:
-
Premium-level SLAs
-
Dedicated support team
-
Email support (Prioritized)
-
Community support
Compare Our ASR Models
Unmatched accuracy. Blazing fast. Most affordable.
Features and Capabilities
Batch processing (1hr of audio)
Audio streaming processing lag
Word Error Rate (WER)
Language detection
Word-level timestamps
Diarization (separate per speaker)
Smart formatting
Deepgram Nova
12.1s
<300ms
9.5%
Median WER across multiple domains using real-world data
Up to 10
OpenAI Whisper:
Large
- Tiny
- Base
- Small
- Medium
- Large
Fully Managed by Deepgram
Calculate to see which plan is right for you.
Select Audio Type:
Select Model:
Deepgram Nova
Pricing: $0.0044 per min
Deepgram Nova
Pricing: $0.0044 per min
Whisper Large
Pricing: $0.0048 per min
Whisper Medium
Pricing: $0.0042 per min
Whisper Small
Pricing: $0.0038 per min
Whisper Base
Pricing: $0.0035 per min
Whisper Tiny
Pricing: $0.0033 per min
Monthly Audio Volume:
0
minutesAdd on:
Audio intelligence Pricing: $0.0043 per minute
The add on description for the Audio Intelligence model should go here to give a bit more depth into the specified features listed below.
- Speaker Diarization
- Entity Detection
- Summarization
- Topic Detection
- Language Translation
- Language Detection
- Sentiment Analysis
Your Suggested Plan:
The best plan for you is:
Pay As You Go
Your rate is:
$0.0000/min
Monthly Estimate:
$0
Additional features you get with the Pay As You Go plan:
- No rate limits
- Prioritized Requests
- Faster Response Times
- Priority Access to New Features
Additional features you get with the Growth plan:
- No rate limits
- Prioritized Requests
- Faster Response Times
- Priority Access to New Features
Exclusive features only found on our Premium plan:
- Small Feature
- Medium Size Feature
- Longer Feature Name Here
- Even Longer Feature Name Goes Here
- Medium Size Feature Here
FAQs
Can I sign up for free?
Absolutely. We’ll even give you free credits to try out our transcription as well as our formatting and understanding features. No credit card required.
What’s the difference between Base and Enhanced model tiers?
Our Base models are built on our signature end-to-end deep learning speech model architecture and offer a solid combination of accuracy and cost-effectiveness. Our Enhanced models generally have even higher accuracy and handle uncommon words significantly better.
Do you offer volume discounts?
Our Growth plan offers 20% savings for pre-paying for credits. If you’re looking to transcribe over 10,000 hours of audio per year, you can save even more with a Premium plan. Contact us for more information.
Which file types can you transcribe?
We support over 40 audio and video formats, documented here.
How does billing work?
You can purchase credits upfront with a credit card. Credits will be deducted from your balance as you use our API. Pay As You Go credits never expire. Growth plan credits expire 1 year from purchase unless you renew or upgrade.
Can you transcribe live streaming audio?
Definitely. In fact, we’ve got the fastest real-time transcription in the biz with latency times of under 300 milliseconds.
What happens if I run out of credits before my plan expires?
If you’re on the Growth plan and have saved a credit card, you can continue to use our API with a 10% overage fee billed at the start of each month.
What languages do you support?
We support over 30 languages and dialects for transcription (see list here) with over 100 supported for translation.
What happens if I have unused credits when my plan expires?
Credits purchased on a Growth plan expire a year from purchase unless you renew or upgrade.
Can I get human support?
Sure thing. You can get help from our community over at Github Discussions or email our support team at [email protected].
How many seats do I get?
We bill based on usage not users. Add as many team members and collaborators as you wish!
Can I deploy Deepgram on-premises or in a VPC?
You sure can. Contact us about getting on a Premium plan to expand your deployment capabilities.
Can I sign up for free?
Absolutely. We’ll even give you free credits to try out our transcription as well as our formatting and understanding features. No credit card required.
Do you offer volume discounts?
Our Growth plan offers 20% savings for pre-paying for credits. If you’re looking to transcribe over 10,000 hours of audio per year, you can save even more with a Premium plan. Contact us for more information.
How does billing work?
You can purchase credits upfront with a credit card. Credits will be deducted from your balance as you use our API. Pay As You Go credits never expire. Growth plan credits expire 1 year from purchase unless you renew or upgrade.
What happens if I run out of credits before my plan expires?
If you’re on the Growth plan and have saved a credit card, you can continue to use our API with a 10% overage fee billed at the start of each month.
What happens if I have unused credits when my plan expires?
Credits purchased on a Growth plan expire a year from purchase unless you renew or upgrade.
How many seats do I get?
We bill based on usage not users. Add as many team members and collaborators as you wish!
What’s the difference between Base and Enhanced model tiers?
Our Base models are built on our signature end-to-end deep learning speech model architecture and offer a solid combination of accuracy and cost-effectiveness. Our Enhanced models generally have even higher accuracy and handle uncommon words significantly better.
Which file types can you transcribe?
We support over 40 audio and video formats, documented here.
Can you transcribe live streaming audio?
Definitely. In fact, we’ve got the fastest real-time transcription in the biz with latency times of under 300 milliseconds.
What languages do you support?
We support over 30 languages and dialects for transcription (see list here) with over 100 supported for translation.
Can I get human support?
Sure thing. You can get help from our community over at Github Discussions or email our support team at [email protected].
Can I deploy Deepgram on-premises or in a VPC?
You sure can. Contact us about getting on a Premium plan to expand your deployment capabilities.