Compare speech-to-text API pricing
Pricing calculator
Stop overpaying for speech AI
1,000 hours of audio / month
Pricing assumptions
Based on public pay-as-you-go pricing. Enterprise discounts and committed-use contracts may differ. Some providers charge separately for certain features. The calculator uses the public price for the provider configuration that most closely matches Soniox.
What each provider charges
Hourly estimates based on each provider’s public pay-as-you-go pricing, converted from tokens, characters, or requests where needed. These are estimates, not quotes. Enterprise, committed-use, and separately billed features may differ.
| Provider | Real-time per hour | Async per hour |
|---|---|---|
| Soniox Updated June 2026 | $0.12 stt-rt-v5 | $0.10 stt-async-v5 |
| Inworld Updated September 2026 | $0.15 inworld-stt-1 | $0.15 inworld-stt-1 |
| Meta Updated September 2026 | $0.18 muse-voice-transcribe-1.0 | $0.18 muse-voice-transcribe-1.0 |
| xAI Updated September 2026 | $0.20 grok-voice-transcribe-2.0 | $0.10 grok-voice-transcribe-2.0 |
| Smallest AI Updated September 2026 | $0.24 Pulse | $0.18 Pulse |
| ElevenLabs Updated September 2026 | $0.39 Scribe v2 Realtime | $0.27 Scribe v2 |
| Cartesia Updated July 2026 | $0.43 ink-2 | $0.43 ink-2 |
| Google Updated August 2026 | $0.54 gemini-3.5-transcribe-live | $0.30 gemini-3.5-transcribe |
| Speechmatics Updated July 2026 | $0.54 Real-time Enhanced Accuracy | $0.50 Batch Enhanced Accuracy |
| AssemblyAI Updated September 2026 | $0.62 Universal-3.5 Pro Realtime | $0.33 Universal-3.5 Pro |
| Deepgram Updated September 2026 | $0.75 Nova-3 Multilingual Streaming | $0.39 Nova-3 Multilingual Batch |
| OpenAI Updated September 2026 | $1.02 gpt-realtime-whisper | $0.36 gpt-4o-transcribe |
| Azure Updated September 2026 | $1.60 Standard Real-time Transcription | $0.18 Standard Batch Transcription |
Prices are each provider’s base hourly rate plus the paid add-ons needed to match features Soniox includes: Deepgram (keyterm prompting, speaker diarization), AssemblyAI (keyterm prompting, general prompting, speaker diarization), ElevenLabs (keyterm prompting), and Azure (continuous language identification, speaker diarization).