AssemblyAI
Developer platform for file, streaming and short-clip transcription, speech analysis, LLM access and managed voice agents.
Category: Audio & Voice · Audio editing & restoration
Features
AssemblyAI provides APIs for prerecorded files, live streams and short synchronous requests. Universal-2 supports 99 languages and Universal-3.5 Pro 18. The Sync API processes up to 2 minutes per request. Diarization, Medical Mode, content analysis and redaction have API-specific extra prices.
The Voice Agent API bundles Universal-3.5 Pro Realtime, its own conversation model and speech synthesis, turn handling, recordings and hosting. You can connect your own Twilio/SIP service; carrier fees and additional Gateway calls are not paid by that bundle. Realtime and Sync are also offered for self-hosting; your own infrastructure is still required.
- File transcription, realtime streaming and Sync API for short clips
- Free starter credits, not an ongoing unlimited free API plan
- Free: 5 new streams/minute; PAYG starts at 100 and scales automatically
- Diarization, Medical Mode, summaries and PII redaction with separate rates
- LLM Gateway with model-specific token billing; SSO is a paid add-on
Prices and plans
| Tier | Price | Period | What you get |
|---|---|---|---|
| Free Credits | free | one-time starter credit | USD 50 credit without expiry; no card required to start For file/realtime STT, voice agents, Speech Understanding and Guardrails; no LLM Gateway 5 new streaming connections/minute; no training opt-out, TTL or BAA on free accounts |
| Pre-recorded / Universal-2 | 0.15 USD | per audio hour; prorated to seconds | 99 languages; keyterms included Extra per audio hour: diarization USD 0.02, Medical Mode USD 0.15 |
| Pre-recorded / Universal-3.5 Pro | 0.21 USD | per audio hour; prorated to seconds | 18 languages and code switching; wider coverage through Universal-2 fallback Extra per audio hour: keyterms USD 0.05, prompting USD 0.05, diarization USD 0.02, Medical Mode USD 0.15 |
| Realtime / Universal-Streaming English | 0.15 USD | per hour of open connection | English; idle time is billed Extra per connection hour: keyterms USD 0.04, diarization USD 0.12, Medical Mode USD 0.15 |
| Realtime / Universal-Streaming Multilingual | 0.15 USD | per hour of open connection | 6 languages: English, Spanish, German, French, Portuguese, Italian Extra per connection hour: keyterms USD 0.04, diarization USD 0.12, Medical Mode USD 0.15 |
| Realtime / Universal-3.5 Pro | 0.45 USD | per hour of open connection | 18 languages; keyterms included, idle time billed Extra per connection hour: diarization USD 0.12, prompting USD 0.05, Medical Mode USD 0.15, Voice Focus USD 0.10 |
| Sync API | 0.45 USD | per audio hour; prorated to seconds | Up to 2 minutes per request, 18 languages, word timing and confidence Keyterms and conversation context included; prompting adds USD 0.05 per audio hour |
| Voice Agent API | 4.50 USD | per hour of open connection | Equivalent to USD 0.075/minute, billed per second; own STT/LLM/TTS stack with recordings and hosting Prompting, Voice Focus, turn/interruption detection included; your Twilio fees and additional Gateway calls separate |
| Speaker Identification / Low effort | 0.02 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| Speaker Identification / Medium effort | 0.10 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| Translation | 0.06 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| Custom Formatting | 0.03 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| Entity Detection | 0.08 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| Sentiment Analysis | 0.02 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| Key Phrases | 0.01 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| Topic Detection | 0.15 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| Summarization / Low effort | 0.02 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| Summarization / Medium effort | 0.07 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| Profanity Filtering | 0.01 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| PII Audio Redaction | 0.05 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| PII Text Redaction | 0.08 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| Content Moderation | 0.15 USD | per audio hour; prorated to seconds | Add-on; transcription base rate applies separately |
| LLM Gateway / GPT-5 Nano | 0.05 / 0.40 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / GPT OSS 20B | 0.07 / 0.30 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Gemini 2.5 Flash Lite | 0.10 / 0.40 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Qwen3.5 4B Fast | 0.10 / 0.50 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / gemma-4-31b | 0.14 / 0.40 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / GPT OSS 120B | 0.15 / 0.60 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Qwen3 32B | 0.15 / 0.60 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Qwen3 Next 80B A3B | 0.15 / 1.20 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Gemini 3.1 Flash Lite | 0.25 / 1.50 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / GPT-5 mini | 0.25 / 2.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Gemini 2.5 Flash | 0.30 / 2.50 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Gemini 3.5 Flash Lite | 0.30 / 2.50 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Gemini 3.7 Flash | 0.75 / 3.75 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Gemini 3.8 Flash | 0.75 / 3.75 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / GPT-5.6 Luna | 1.00 / 6.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Haiku 4.5 | 1.00 / 5.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Gemini 2.5 Pro | 1.25 / 10.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Gemini 3.5 Flash | 1.25 / 9.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / GPT-5 | 1.25 / 10.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / GPT-5.1 | 1.25 / 10.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Gemini 3.6 Flash | 1.50 / 7.50 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / GPT-5.2 | 1.75 / 14.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / GPT-4.1 | 2.00 / 8.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / GPT-5.6 Terra | 2.50 / 15.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Sonnet 4.5 | 3.00 / 15.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Sonnet 4.6 | 3.00 / 15.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Sonnet 5 | 3.00 / 15.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / GPT-5.6 Sol | 4.00 / 20.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / GPT-5.5 | 5.00 / 30.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Opus 4.5 | 5.00 / 25.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Opus 4.6 | 5.00 / 25.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Opus 4.7 | 5.00 / 25.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Opus 4.8 | 5.00 / 25.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| LLM Gateway / Opus 5 | 5.00 / 25.00 USD | per 1M input/output tokens | LLM Gateway, global routing (model_region=global); US/EU regions +10% Not covered by free credits; additional audio/voice sessions billed separately |
| SSO | 199 USD | per connection/month | SAML/OIDC on PAYG and Contract; maximum 1 connection per organization, no SCIM Full month charged on activation, no prorated refund; PAYG needs sufficient balance and enabled auto-pay |
| Custom / Enterprise | contact sales | custom quote | Volume rates, agreed limits and support; contractual scope |
| Self-hosted / Realtime + Sync | contact sales | custom quote | Deployment on your own infrastructure; Realtime Universal-3.5 Pro and Sync API Vendor states the same usage-based cloud rates without a self-hosting premium; clarify infrastructure costs and deployment separately |
Limits
Upload recordings only with the necessary rights and consent. AssemblyAI may use contract-permitted files for its own training after PII redaction, except on EU servers, under a BAA or with opt-out enabled. Current Data Controls are paid-account features offering opt-out, TTL and BAA at no additional charge; changes apply to new requests. Older support text still describes email opt-out and loss of discounts. Streaming zero retention requires training opt-out and does not remove billing metadata. Files, transcripts and Gateway outputs have different retention rules. Streaming scaling raises the next minute's limit by 10% at 70% or greater utilization; this does not promise arbitrary free-tier concurrency.
Similar tools
FAQ
Is AssemblyAI available for free?
Not permanently. What is published is a one-off free allowance.
Where is AssemblyAI's vendor based?
AssemblyAI, Inc., based in USA. For data processing and privacy terms the vendor's own contract governs.
How is AssemblyAI billed?
Vendor USD prices, no currency conversion. Files are billed to the second per audio channel; streaming and voice agents use WebSocket-open time including idle periods. Without session.end, voice agents retain a billable 30-second resume window. Additional LLM Gateway calls cost extra. Gateway rates apply to global routing with model_region=global; regional US/EU rates are 10% higher. The pricing FAQ describes monthly invoices, whereas current self-service billing docs describe prepaid funds and enterprise invoicing. Check the chosen account's terms.
Is AssemblyAI still active?
Active developer API with usage-based billing.
How many plans are documented for AssemblyAI?
The catalog documents 59 plans with price, billing period, and included features.
Which use case is AssemblyAI categorized for?
AssemblyAI is listed in the catalog under “Audio & Voice.”
Where do the details about AssemblyAI come from?
Prices, limits, and product details are documented against the listed official vendor page.
How we compare
We compare free access, paid entry, plan limits, features and the listed official sources. Unsourced figures stay marked as not published.
Only confirmed entries appear in the catalog. Prices, plans and limits come from listed official vendor pages. Missing figures stay marked as not published. Result lists are alphabetical by default and can be switched to Free access first; neither is a quality ranking. The vendor page always wins.
Edited and checked by the AI Compare team
Sources
- https://www.assemblyai.com/pricing
- https://www.assemblyai.com/docs/billing-and-pricing
- https://www.assemblyai.com/docs/getting-started/models
- https://www.assemblyai.com/docs/data-controls
- https://www.assemblyai.com/docs/data-retention-and-model-training
- https://support.assemblyai.com/articles/5930031898-how-to-opt-out-of-data-sharing-for-model-training
- https://www.assemblyai.com/docs/sso
- https://www.assemblyai.com/deployments/self-hosted
Visit site (opens in a new tab)
Spotted an outdated figure? Vendor pages change without notice. Report an error