Speech to text pricing
Speech-to-text API pricing, per hour of audio
9 providers' transcription prices, each read off the provider's own pricing page with a link on every row. Prices billed per minute are also shown per hour, so they compare.
9
Providers
37
Prices listed
$0.04
Lowest per hour
2026-10-25
Prices checked
Prices
Speech to text prices, every provider
Sorted by price per hour of audio. Pre-recorded and streaming are separate rows where the provider prices them apart.
| Model | Provider | Price as listed | Per hour | Notes | Source |
|---|---|---|---|---|---|
| Whisper Large V3 Turbo | Groq | $0.04 per hour | $0.04 | Minimum 10 seconds billed per request. | Sourcechecked 2026-10-25 |
| NVIDIA Parakeet TDT 0.6B v3 | Together AI | $0.0015 per minute | $0.09 | Sourcechecked 2026-10-25 | |
| Whisper Large v3 | Together AI | $0.0015 per minute | $0.09 | Sourcechecked 2026-10-25 | |
| Speech to Text (REST) | xAI | $0.10 per hour | $0.10 | Sourcechecked 2026-10-25 | |
| Whisper (whisper-large-v3) | Groq | $0.111 per hour | $0.111 | Minimum 10 seconds billed per request. | Sourcechecked 2026-10-25 |
| Universal-2pay as you go, pre-recorded | AssemblyAI | $0.15 per hour | $0.15 | 99 languages | Sourcechecked 2026-10-25 |
| Universal-Streamingpay as you go, streaming | AssemblyAI | $0.15 per hour | $0.15 | English only | Sourcechecked 2026-10-25 |
| Universal-Streaming Multilingualpay as you go, streaming | AssemblyAI | $0.15 per hour | $0.15 | English, Spanish, German, French, Portuguese, Italian | Sourcechecked 2026-10-25 |
| gpt-4o-mini-transcribefile transcription | OpenAI | $0.003 per minute | $0.18 | Estimated cost. Billed $1.25 in / $5.00 out per 1M tokens. | Sourcechecked 2026-10-25 |
| Speech to Text (Streaming) | xAI | $0.20 per hour | $0.20 | Sourcechecked 2026-10-25 | |
| Universal-3.5 Propay as you go, pre-recorded | AssemblyAI | $0.21 per hour | $0.21 | 18 languages | Sourcechecked 2026-10-25 |
| Voxtral Mini Transcribe 3 | Mistral AI | $0.0035 per minute | $0.21 | Sourcechecked 2026-10-25 | |
| Scribe v2all plans, batch | ElevenLabs | $0.22 per hour | $0.22 | 90+ languages | Sourcechecked 2026-10-25 |
| Scribe v2 Medicalall plans, batch | ElevenLabs | $0.22 per hour | $0.22 | Clinical audio, HIPAA-eligible | Sourcechecked 2026-10-25 |
| Voxtral Small Transcribe 3 | Mistral AI | $0.004 per minute | $0.24 | Sourcechecked 2026-10-25 | |
| Nova-3 Monolingualpay as you go, pre-recorded | Deepgram | $0.0043 per minute | $0.258 | Growth $0.0036/min | Sourcechecked 2026-10-25 |
| gpt-transcribefile transcription | OpenAI | $0.0045 per minute | $0.27 | Sourcechecked 2026-10-25 | |
| GPT-TranscribeGlobal | Microsoft Azure | $0.27 per hour | $0.27 | Sourcechecked 2026-10-25 | |
| Nova-3 Monolingualpay as you go, streaming | Deepgram | $0.0048 per minute | $0.288 | Promo price. Regular price $0.0077/min. Growth $0.0042/min (regular $0.0065/min). | Sourcechecked 2026-10-25 |
| Whisper Largepay as you go, pre-recorded | Deepgram | $0.0048 per minute | $0.288 | Same on Growth | Sourcechecked 2026-10-25 |
| Sync APIpay as you go, sync (clips up to 2 minutes) | AssemblyAI | $0.30 per hour | $0.30 | Universal-3.5 Pro accuracy in one call | Sourcechecked 2026-10-25 |
| Nova-3 Multilingualpay as you go, pre-recorded | Deepgram | $0.0052 per minute | $0.312 | Growth $0.0043/min | Sourcechecked 2026-10-25 |
| Nova-3 Multilingualpay as you go, streaming | Deepgram | $0.0058 per minute | $0.348 | Promo price. Regular price $0.0092/min. Growth $0.0050/min (regular $0.0078/min). | Sourcechecked 2026-10-25 |
| Nova-2pay as you go, streaming (legacy) | Deepgram | $0.35 per hour | $0.35 | Older model, kept for existing deployments | Sourcechecked 2026-10-25 |
| gpt-4o-transcribefile transcription | OpenAI | $0.006 per minute | $0.36 | Estimated cost. Billed $2.50 in / $10.00 out per 1M tokens. | Sourcechecked 2026-10-25 |
| gpt-4o-transcribe-diarizefile transcription with diarization | OpenAI | $0.006 per minute | $0.36 | Estimated cost. Billed $2.50 in / $10.00 out per 1M tokens. | Sourcechecked 2026-10-25 |
| Whisperfile transcription | OpenAI | $0.006 per minute | $0.36 | Sourcechecked 2026-10-25 | |
| Flux Englishpay as you go, streaming | Deepgram | $0.0065 per minute | $0.39 | Promo price. Regular price $0.0077/min. Growth $0.0057/min (regular $0.0065/min). | Sourcechecked 2026-10-25 |
| Scribe v2 Realtimeall plans, streaming | ElevenLabs | $0.39 per hour | $0.39 | Sourcechecked 2026-10-25 | |
| Whisperus-east | Microsoft Azure | $0.396 per hour | $0.396 | About $0.0066 per minute. | Sourcechecked 2026-10-25 |
| Universal-3.6 Pro Realtimepay as you go, streaming | AssemblyAI | $0.45 per hour | $0.45 | 32 languages | Sourcechecked 2026-10-25 |
| Flux Multilingualpay as you go, streaming | Deepgram | $0.0078 per minute | $0.468 | Growth $0.0068/min | Sourcechecked 2026-10-25 |
| Voxtral Mini Transcribe Realtime 3 | Mistral AI | $0.008 per minute | $0.48 | Realtime / streaming. | Sourcechecked 2026-10-25 |
| Dictation APIpay as you go | AssemblyAI | $0.62 per hour | $0.62 | Cleaned-up text from one utterance, clips up to 120 seconds | Sourcechecked 2026-10-25 |
| gpt-live-transcribelive transcription | OpenAI | $0.017 per minute | $1.02 | Sourcechecked 2026-10-25 | |
| gpt-realtime-whisperlive transcription | OpenAI | $0.017 per minute | $1.02 | Sourcechecked 2026-10-25 | |
| gpt-realtime-translatelive translation | OpenAI | $0.034 per minute | $2.04 | Sourcechecked 2026-10-25 |
Estimator
Estimate a month of speech to text
Enter your monthly volume. Every row that bills in the same unit is priced, cheapest first.
Write 40, 250 or 5k.
100 hours of audio a month costs from $4.00 (Whisper Large V3 Turbo) to $204.00 (gpt-realtime-translate).
| Model | Provider | Per month | Per hour |
|---|---|---|---|
| Whisper Large V3 Turbo | Groq | $4.00 | $0.04 |
| NVIDIA Parakeet TDT 0.6B v3 | Together AI | $9.00 | $0.09 |
| Whisper Large v3 | Together AI | $9.00 | $0.09 |
| Speech to Text (REST) | xAI | $10.00 | $0.10 |
| Whisper (whisper-large-v3) | Groq | $11.10 | $0.111 |
| Universal-2pay as you go, pre-recorded | AssemblyAI | $15.00 | $0.15 |
| Universal-Streamingpay as you go, streaming | AssemblyAI | $15.00 | $0.15 |
| Universal-Streaming Multilingualpay as you go, streaming | AssemblyAI | $15.00 | $0.15 |
| gpt-4o-mini-transcribefile transcription | OpenAI | $18.00 | $0.18 |
| Speech to Text (Streaming) | xAI | $20.00 | $0.20 |
| Universal-3.5 Propay as you go, pre-recorded | AssemblyAI | $21.00 | $0.21 |
| Voxtral Mini Transcribe 3 | Mistral AI | $21.00 | $0.21 |
| Scribe v2all plans, batch | ElevenLabs | $22.00 | $0.22 |
| Scribe v2 Medicalall plans, batch | ElevenLabs | $22.00 | $0.22 |
| Voxtral Small Transcribe 3 | Mistral AI | $24.00 | $0.24 |
| Nova-3 Monolingualpay as you go, pre-recorded | Deepgram | $25.80 | $0.258 |
| gpt-transcribefile transcription | OpenAI | $27.00 | $0.27 |
| GPT-TranscribeGlobal | Microsoft Azure | $27.00 | $0.27 |
| Nova-3 Monolingualpay as you go, streaming | Deepgram | $28.80 | $0.288 |
| Whisper Largepay as you go, pre-recorded | Deepgram | $28.80 | $0.288 |
| Sync APIpay as you go, sync (clips up to 2 minutes) | AssemblyAI | $30.00 | $0.30 |
| Nova-3 Multilingualpay as you go, pre-recorded | Deepgram | $31.20 | $0.312 |
| Nova-3 Multilingualpay as you go, streaming | Deepgram | $34.80 | $0.348 |
| Nova-2pay as you go, streaming (legacy) | Deepgram | $35.00 | $0.35 |
| gpt-4o-transcribefile transcription | OpenAI | $36.00 | $0.36 |
| gpt-4o-transcribe-diarizefile transcription with diarization | OpenAI | $36.00 | $0.36 |
| Whisperfile transcription | OpenAI | $36.00 | $0.36 |
| Flux Englishpay as you go, streaming | Deepgram | $39.00 | $0.39 |
| Scribe v2 Realtimeall plans, streaming | ElevenLabs | $39.00 | $0.39 |
| Whisperus-east | Microsoft Azure | $39.60 | $0.396 |
| Universal-3.6 Pro Realtimepay as you go, streaming | AssemblyAI | $45.00 | $0.45 |
| Flux Multilingualpay as you go, streaming | Deepgram | $46.80 | $0.468 |
| Voxtral Mini Transcribe Realtime 3 | Mistral AI | $48.00 | $0.48 |
| Dictation APIpay as you go | AssemblyAI | $62.00 | $0.62 |
| gpt-live-transcribelive transcription | OpenAI | $102.00 | $1.02 |
| gpt-realtime-whisperlive transcription | OpenAI | $102.00 | $1.02 |
| gpt-realtime-translatelive translation | OpenAI | $204.00 | $2.04 |
List prices in USD on 2026-10-25, before tax, free tiers and add-ons. Rows billed another way are in the price table only.
How to choose
How to choose on price
Pre-recorded or streaming
If nobody needs the words while the call is live, upload files after. On every provider that lists both, it costs the same or less.
Count channels, not calls
Stereo and multichannel audio is often billed per channel. Mix down to mono when you do not need to tell speakers apart by channel.
Price the add-ons
Speaker labels, key terms and redaction are extra on several providers. A cheap base rate with three add-ons can cost more than an all-in rate.
Test on your own audio
Accents, background noise, jargon and phone-line audio change accuracy a lot. Run the same hour through two or three providers before you choose.
Watch promotions
Some streaming prices here are promotional. Plan the budget on the regular price in the row note.
Check the minimums
Some providers bill a minimum per request, such as 10 seconds. Many short clips can cost more than the per-hour price suggests.
FAQ
Speech to text pricing questions
Which speech-to-text API is cheapest?
By list price, Whisper Large V3 Turbo from Groq at $0.04 per hour of audio, checked 2026-10-25. The dearest row here is gpt-realtime-translate at $2.04 per hour. Accuracy on your own audio matters more than the gap, so test a few.
Why are streaming prices higher than pre-recorded?
Streaming returns words while the person is still talking, which costs the provider more to serve. Most providers here list a higher price for streaming or realtime models than for files you upload after the call.
What adds to the per-hour price?
Add-ons such as speaker labels, key-term prompting, redaction and medical mode cost extra on several providers. Stereo or multichannel audio is often billed per channel, so a one-hour stereo call can bill as two hours. Each vendor page lists the add-ons its pricing page shows.
How current are these prices?
Every row was read off the provider's own pricing page on 2026-10-25 and links to it. Deepgram marks some streaming prices as a limited-time promotion; the row notes give the regular price.
Related
Vendor pages and tools
OpenAI API pricing10 models and 20 other prices, checked 2026-10-25.Azure OpenAI pricing9 models and 7 other prices, checked 2026-10-25.Groq API pricing4 models and 4 other prices, checked 2026-10-25.Mistral API pricing9 models and 6 other prices, checked 2026-10-25.Together AI API pricing10 models and 11 other prices, checked 2026-10-25.Grok API pricing7 models and 6 other prices, checked 2026-10-25.AssemblyAI API pricing7 prices, checked 2026-10-25.Deepgram API pricing11 prices, checked 2026-10-25.ElevenLabs API pricing9 prices, checked 2026-10-25.Voice agent costSpeech to text, an LLM and text to speech, priced per call.Embeddings pricingEvery embeddings price we track.Image generation pricingEvery image generation price we track.Text to speech pricingEvery text to speech price we track.Vector databases pricingEvery vector databases price we track.LLM API pricing comparedToken prices for every LLM vendor we track.All pricing pagesEvery vendor we track, with the day each was checked.
Building this into a product? OceanLab builds AI into business workflows, with people kept in charge. AI automation →
