Speech / TTS · Updated 2026-10-25
AssemblyAI — what it is actually for
AssemblyAI provides pre-recorded and streaming speech-to-text, audio intelligence features, and a voice agent API. Pricing is pay-as-you-go per audio hour, with add-ons priced per hour.
Short answer: use AssemblyAI when you need transcription with speaker diarization. It is a free credits + usage-based api service from AssemblyAI; the official pricing source is linked below.
- Vendor
- AssemblyAI
- Category
- Speech / TTS
- Pricing model
- Free credits + usage-based API
- Free option
- Yes. New accounts receive US$50 in free credits according to AssemblyAI.
What AssemblyAI is used for
- Transcription with speaker diarization
- PII redaction and audio intelligence
- Voice agents that combine speech-to-text, LLM, and TTS
Source: official product page and official documentation.
Who AssemblyAI is for
- Developers who need transcription plus audio intelligence
- High-concurrency transcription workloads
- Teams that want predictable per-hour costs
Source: official documentation.
Who it is not for
- Users who want a consumer transcription app
- Buyers who need unlimited free usage
Source: documented plans, limits, and deployment model on the official product page.
Pricing
- US$50 in free credits, then pay-as-you-go with no minimum commitment.Source: www.assemblyai.com
- Universal-2 asynchronous speech-to-text is listed at US$0.15/hour; streaming is listed at US$0.15/hour.Source: www.assemblyai.com
- The Voice Agent API is listed at US$4.50/hour (US$0.075/minute). Optional features such as diarization and PII redaction are priced per hour.Source: www.assemblyai.com
Pre-recorded audio is billed per audio hour submitted; multichannel audio is billed per channel.
Full pricing page: https://www.assemblyai.com/pricing
Alternatives to AssemblyAI
- Deepgram — Speech-to-text, text-to-speech, and voice agent APIs with per-minute billing.
- OpenAI Audio API — Speech-to-text and text-to-speech APIs billed by usage.
- Azure AI Speech — Enterprise speech-to-text, neural TTS, translation, and speaker recognition.
Official sources
AssemblyAI FAQ
Is AssemblyAI free?
New accounts receive US$50 in free credits. After that, usage is billed pay-as-you-go.
How much does AssemblyAI transcription cost?
The official pricing page lists Universal-2 asynchronous speech-to-text at US$0.15/hour and streaming at US$0.15/hour.
Does AssemblyAI include diarization?
Speaker diarization is available, with per-hour add-on pricing for asynchronous audio according to the pricing page.