Meetings
Minutes of stand-ups and calls.
> Transcribe the call and pull out the tasksMeetings, interviews, lectures, and voice messages — turned into text in minutes. Pay via SBP, and pay only for minutes of audio.
> you: file: interview.m4a · 18 min
scribe-v2 ▸
[ok] 18 min of audio transcribed · charged $0.13
The aggregator has 26 speech transcription models:
Go by the task: there's no need to pay for the most expensive model where a simpler one will do.
Minutes of stand-ups and calls.
> Transcribe the call and pull out the tasksText for journalists and researchers.
> Transcribe a 40-minute interviewSummary in text.
> Transcribe the webinarLong voice messages — into text.
> Turn a voice message into textNo subscription: pay only for transcription. The price is from $0.0004 to $0.032 per minute of audio.
Beta 1.0 prices may change
| Model | Developer | Minute of audio | Hour of audio | $10 is enough for |
|---|---|---|---|---|
| Universal-3.6 Pro | AssemblyAI | $0.01 | $0.60 | ≈ 16 h |
| Scribe v2 | ElevenLabs | $0.0073 | $0.44 | ≈ 22 h |
| Scribe v2 Medical | ElevenLabs | $0.0073 | $0.44 | ≈ 22 h |
| Gemini 3.5 Transcribe | $0.011 | $0.65 | ≈ 15 h | |
| Transcribe 1 Pro | Fish Audio | $0.012 | $0.72 | ≈ 13 h |
| Universal-3.5 Pro | AssemblyAI | $0.01 | $0.60 | ≈ 16 h |
| Muse Voice Transcribe 1.0 | Meta | $0.006 | $0.36 | ≈ 27 h |
| MAI-Transcribe 2 | Microsoft AI | $0.0033 | $0.20 | ≈ 50 h |
| Nemotron 3.5 ASR Streaming Multilingual 0.6B | NVIDIA | $0.0004 | $0.024 | ≈ 416 h |
| Voxtral Small 24B 2507 STT | Mistral | $0.006 | $0.36 | ≈ 27 h |
| Voxtral Mini 3B 2507 | Mistral | $0.002 | $0.12 | ≈ 83 h |
| Qwen3 ASR 1.7B | Qwen | $0.0009 | $0.054 | ≈ 185 h |
| Qwen3 ASR 0.6B | Qwen | $0.0004 | $0.024 | ≈ 416 h |
| GPT Transcribe | OpenAI | $0.009 | $0.54 | ≈ 18 h |
| Transcribe 1 | Fish Audio | $0.012 | $0.72 | ≈ 13 h |
| Grok STT 1.0 | SpaceXAI | $0.0033 | $0.20 | ≈ 50 h |
| Nova-3 | Deepgram | $0.0086 | $0.52 | ≈ 19 h |
| MAI-Transcribe 1.5 | Microsoft AI | $0.012 | $0.72 | ≈ 13 h |
| Parakeet TDT 0.6B v3 | NVIDIA | $0.003 | $0.18 | ≈ 55 h |
| Voxtral Mini Transcribe | Mistral | $0.006 | $0.36 | ≈ 27 h |
| Chirp 3 | $0.032 | $1.92 | ≈ 5 h | |
| GPT-4o Mini Transcribe | OpenAI | $0.0057 | $0.34 | ≈ 28 h |
| Whisper Large V3 | OpenAI | $0.0009 | $0.054 | ≈ 185 h |
| Whisper Large V3 Turbo | OpenAI | $0.0004 | $0.024 | ≈ 416 h |
| Whisper 1 | OpenAI | $0.012 | $0.72 | ≈ 13 h |
| GPT-4o Transcribe | OpenAI | $0.011 | $0.69 | ≈ 14 h |
The price is per minute of audio recording. The actual transcription cost is deducted from your balance.
One account opens all the aggregator's models — no need to register with each one separately.
Balance in dollars. No foreign card or VPN needed, no monthly fee.
Open the AI aggregator, choose a model, and add an audio recording. The cost of the result is deducted from your balance.
From $0.024 to $1.92 per hour of recording — depends on the model. For example, Whisper Large V3 Turbo costs about $0.024 per hour.
Yes, Whisper, GPT Transcribe, Scribe v2, and other multilingual models recognize Russian speech.
No. The P2D aggregator opens from Russia without a VPN, and the service itself sends the requests to the models.
Top up your balance in dollars via SBP with a Russian card. No foreign card needed, and there's no subscription or monthly fee — you're charged only for what you generate.
Sign up in a minute, top up via SBP, pay only for the result.