AssemblyAI pricing per hour and per minute
AssemblyAI's transcription costs $0.15 to $0.30 an hour for recorded audio and $0.15 to $0.45 an hour for live audio.
What the numbers say
- For recorded audio at 100 hours a month, AssemblyAI's Universal-2 costs $0.15 an hour, 17% below the median of the 11 vendors ($0.18): 3rd, after Speechmatics at $0.12 and ahead of OpenAI at $0.18.
- Its place for recorded audio moves from 4th at 10 hours a month to 3rd at 10,000 hours, passing ElevenLabs.
- For live audio at 100 hours a month, AssemblyAI's Universal-Streaming costs $0.15 an hour, 72% below the median of the 9 vendors ($0.536): 1st, ahead of Speechmatics at $0.20.
- Live audio costs the same as recorded audio on AssemblyAI's cheapest models; across the 9 vendors that price both, the median premium is 75%.
Prices by model
| Model | Audio | Per minute | Per hour |
|---|---|---|---|
| Universal-2 | batch | $0.0025 | $0.15 |
| AssemblyAI-STT-3.7 | batch | $0.0035 | $0.21 |
| Sync API Flagship accuracy in one call, for clips up to 20 minutes. | batch | $0.005 | $0.30 |
| Universal-Streaming English only. Billed on connection time, not audio sent. No longer on the pricing page, which lists the AssemblyAI-STT-3.7 models; the pricing reference still lists it at this price. | streaming | $0.0025 | $0.15 |
| Universal-Streaming Multilingual Billed on connection time, not audio sent. No longer on the pricing page, which lists the AssemblyAI-STT-3.7 models; the pricing reference still lists it at this price. | streaming | $0.0025 | $0.15 |
| AssemblyAI-STT-3.7-Realtime Billed on connection time, not audio sent. | streaming | $0.005 | $0.30 |
| AssemblyAI-STT-3.7-Realtime-Agent Billed on connection time, not audio sent. | streaming | $0.0075 | $0.45 |
What 10, 100, 1,000 and 10,000 hours a month cost
| Model | 10 hours | 100 hours | 1,000 hours | 10,000 hours |
|---|---|---|---|---|
| Universal-2 batch | $1.50 $0.15/h | $15 $0.15/h | $150 $0.15/h | $1,500 $0.15/h |
| AssemblyAI-STT-3.7 batch | $2.10 $0.21/h | $21 $0.21/h | $210 $0.21/h | $2,100 $0.21/h |
| Sync API batch | $3 $0.30/h | $30 $0.30/h | $300 $0.30/h | $3,000 $0.30/h |
| Universal-Streaming streaming | $1.50 $0.15/h | $15 $0.15/h | $150 $0.15/h | $1,500 $0.15/h |
| Universal-Streaming Multilingual streaming | $1.50 $0.15/h | $15 $0.15/h | $150 $0.15/h | $1,500 $0.15/h |
| AssemblyAI-STT-3.7-Realtime streaming | $3 $0.30/h | $30 $0.30/h | $300 $0.30/h | $3,000 $0.30/h |
| AssemblyAI-STT-3.7-Realtime-Agent streaming | $4.50 $0.45/h | $45 $0.45/h | $450 $0.45/h | $4,500 $0.45/h |
Regular prices. One-off credits and promotions aren't counted.
Free tier and credits
- Once, at sign-up: Sign-up comes with $50 of free credit and no card required; AssemblyAI says that covers up to 185 hours of pre-recorded or 333 hours of streaming transcription.
Every vendor's free tier, in hours of audio: free speech-to-text APIs.
Billing
- Pre-recorded audio is billed per hour of audio submitted; streaming is billed for the time the connection is open, idle time included. AssemblyAI doesn't state a rounding increment.
- Multichannel audio is billed per channel: a one-hour stereo file counts as two hours.
- AssemblyAI's own pricing reference lists keyterms on Universal-2 at $0.05 an hour, while its pricing page says included.
Add-ons
| Feature | Audio | Price |
|---|---|---|
| Speaker diarization | batch | +$0.02 an hour |
| Speaker diarization | streaming | +$0.12 an hour |
| Keyterm prompting | batch | +$0.05 an hour AssemblyAI-STT-3.7; the pricing page lists it as included on Universal-2. |
| Medical mode | batch, streaming | +$0.15 an hour |
| PII redaction | batch | +$0.08 an hour Redacts the transcript text; redacting the audio itself is $0.05 an hour. |
| Entity detection | batch | +$0.08 an hour |
Volume discounts and commitments
- Volume discounts are custom, through sales.
Also priced, not ranked
Specialist products and promotion-only prices, kept out of the comparisons:
- Dictation API (batch)
- $0.0103 a minute ($0.62 an hour). Returns cleaned-up text from clips of up to 120 seconds.
Quick answers
How is AssemblyAI priced?
AssemblyAI charges $0.15 to $0.30 an hour for recorded audio ($0.0025 to $0.005 a minute) across 3 options, and $0.15 to $0.45 an hour for live audio ($0.0025 to $0.0075 a minute) across 4 options. Sign-up comes with $50 of free credit and no card required; AssemblyAI says that covers up to 185 hours of pre-recorded or 333 hours of streaming transcription. Pre-recorded audio is billed per hour of audio submitted; streaming is billed for the time the connection is open, idle time included. AssemblyAI doesn't state a rounding increment.
How much does AssemblyAI cost per hour?
$0.15 to $0.30 an hour for recorded audio, and $0.15 to $0.45 an hour for live audio, at regular pay-as-you-go prices.
What do 100 hours of audio a month cost on AssemblyAI?
$15 for recorded audio on Universal-2, or $15 for live audio on Universal-Streaming.
Does AssemblyAI have a free tier?
Sign-up comes with $50 of free credit and no card required; AssemblyAI says that covers up to 185 hours of pre-recorded or 333 hours of streaming transcription.
Does AssemblyAI bill per second?
Pre-recorded audio is billed per hour of audio submitted; streaming is billed for the time the connection is open, idle time included. AssemblyAI doesn't state a rounding increment.
Is anything cheaper than AssemblyAI for recorded audio?
Yes: at 100 hours a month, 2 of the 10 other vendors we price cost less: Groq Whisper Large V3 Turbo at $0.04 an hour and Speechmatics Melia 1 at $0.12 an hour.
Is anything cheaper than AssemblyAI for live audio?
No. At 100 hours a month, none of the 8 other vendors we price costs less.
AssemblyAI against the market: recorded audio
| # | Vendor and option | Per month | Per hour | Per minute |
|---|---|---|---|---|
| 1 | Groq Whisper Large V3 Turbo | $4 | $0.04 | $0.0007 |
| 2 | Speechmatics Melia 1 | $12 | $0.12 | $0.002 |
| 3 | AssemblyAI Universal-2 | $15 | $0.15 | $0.0025 |
| 4 | OpenAI gpt-4o-mini-transcribe | $18 | $0.18 | $0.003 |
| 4 | Google Cloud V2 dynamic batch | $18 | $0.18 | $0.003 |
AssemblyAI against the market: live audio
| # | Vendor and option | Per month | Per hour | Per minute |
|---|---|---|---|---|
| 1 | AssemblyAI Universal-Streaming | $15 | $0.15 | $0.0025 |
| 2 | Speechmatics Melia 1 | $20 | $0.20 | $0.0033 |
| 3 | ElevenLabs Scribe v2 Realtime | $21.45 | $0.214 | $0.0036 |
Compare and calculate
- Speech-to-text cost calculatorPrice your own monthly hours
- What it costs to transcribe an hour of audioA single recording
- Cheapest speech-to-text API
- AssemblyAI vs ElevenLabs ScribeHead to head
Sources
2 pages read on 6 October 2026, with 16 passages quoted word for word
- www.assemblyai.com/pricing (plain HTTP)
- www.assemblyai.com/pricing.md (plain HTTP; AssemblyAI's own pricing reference, for billing rules)
The quoted text
Models Pay as you go AssemblyAI-STT-3.7 New The most accurate model for recorded audio, across 32 languages with automatic language detection and native code switching, plus a new speaker diarization model that holds up on short clips. $0.21 /hr Universal-2 Our highly accurate speech-to-text model trained on over 12.5 million hours of audio data. Supports 99 languages. Exceptional accuracy at a lower price. $0.15 /hr
www.assemblyai.com/pricing
Add-on features AssemblyAI-STT-3.7 Universal-2 Keyterms Prompting Provide up to 1000 words or phrases (maximum 6 words per phrase) to improve transcription accuracy. $0.05 /hr Included Prompting Describe your audio in plain language to improve transcription accuracy — cover the domain, scenario, or full conversation details. $0.05 /hr Not supported Speaker Diarization Detect multiple speakers in audio files and segment the transcript into utterances, showing what each speaker said. $0.02 /hr $0.02 /hr Medical Mode New Optimize transcription for medical terminology and healthcare conversations with significantly improved accuracy. $0.15 /hr $0.15 /hr
www.assemblyai.com/pricing
Models Pay as you go AssemblyAI-STT-3.7-Realtime New For notetakers and ambient transcription, with built-in keyterm prompting, context carryover, and conversation memory. Supports 32 languages and realtime speaker labels. $0.30 /hr AssemblyAI-STT-3.7-Realtime-Agent New The most accurate model for voice agents, with built-in keyterm prompting, context carryover, and conversation memory. Supports 32 languages and Voice Focus. $0.45 /hr
www.assemblyai.com/pricing
Speaker Diarization Detect multiple speakers in audio files and segment the transcript into utterances, showing what each speaker said. AssemblyAI-STT-3.7-Realtime-Agent $0.12 /hr AssemblyAI-STT-3.7-Realtime $0.12 /hr
www.assemblyai.com/pricing
Sync API New Flagship transcription accuracy delivered in a single call. Process up to 20 minutes per request, get results back in ~134 ms (p50). Includes per-word timing and confidence across 32 languages. $0.30 /hr
www.assemblyai.com/pricing
Dictation API New Finished text from a single spoken utterance, ready to go straight into whatever your user is writing. Flagship transcription accuracy across 32 languages, with recordings up to 20 minutes per request. The verbatim transcript comes back alongside it. $0.62 /hr
www.assemblyai.com/pricing
PII Audio Redaction Identify and remove PII from the audio file before it is returned to you $0.05 /hr PII Text Redaction Identify and remove PII from the transcription text before it is returned to you $0.08 /hr
www.assemblyai.com/pricing
Entity Detection Identify entities that are spoken, such as names or email addresses $0.08 /hr
www.assemblyai.com/pricing
Yes. You can create an account and start transcribing immediately with no credit card required. The free tier includes up to 185 hours of pre-recorded transcription and up to 333 hours of streaming transcription.
www.assemblyai.com/pricing
**$50 in free credits, no credit card required.**
www.assemblyai.com/pricing.md
Yes. We offer custom pricing for customers with high-volume usage.
www.assemblyai.com/pricing
Multichannel audio is billed per channel. For example, a 1-hour stereo (2-channel) file is billed as 2 hours of transcription.
www.assemblyai.com/pricing
- **Pre-recorded** is billed per **hour of audio submitted**.
www.assemblyai.com/pricing.md
- **Streaming** is billed per **session duration** — the time the WebSocket connection is open, **not** the duration of audio sent. Idle connection time counts.
www.assemblyai.com/pricing.md
| Universal-Streaming English (replaced by AssemblyAI-STT-3.7-Realtime) | `universal-streaming-english` | **$0.15/hr** |
www.assemblyai.com/pricing.md
| Universal-2 (async) | +$0.05/hr | Up to 200 terms |
www.assemblyai.com/pricing.md
Prices on this page were checked against the official pricing pages on .