Hormouz AIHormouzSign in

Speech-to-text (STT) models

Speech-to-text (STT) models on Hormouz. 24 of them, including AssemblyAI: Universal-3.5 Pro. Transcribe audio into text through one API.

  1. 1AssemblyAI: Universal-3.5 ProAssemblyai · audio$0.0075/minute of audio
  2. 2Deepgram: Nova-3Deepgram · audio$0.0052/minute of audio
  3. 3ElevenLabs: Scribe v2Elevenlabs · audio$0.003667/minute of audio
  4. 4ElevenLabs: Scribe v2 MedicalElevenlabs · audio$0.003667/minute of audio
  5. 5Fish Audio: Transcribe 1Fish Audio · audio$0.006/minute of audio
  6. 6Fish Audio: Transcribe 1 ProFish Audio · audio$0.006/minute of audio
  7. 7Google: Chirp 3Google · audio$0.016/minute of audio
  8. 8Google: Gemini 3.5 TranscribeGoogle · audio$2/M audio tokens · $12/M text tokens
  9. 9Microsoft AI: MAI-Transcribe 1.5Microsoft · audio$0.006/minute of audio
  10. 10Microsoft AI: MAI-Transcribe 2Microsoft · audio$0.001667/minute of audio
  11. 11Mistral: Voxtral Mini 3B 2507Mistral AI · audio$0.001/minute of audio
  12. 12Mistral: Voxtral Mini TranscribeMistral AI · audio$0.0033/minute of audio
  13. 13Mistral: Voxtral Small 24B 2507 STTMistral AI · audio$0.003/minute of audio
  14. 14NVIDIA: Nemotron 3.5 ASR Streaming Multilingual 0.6BNVIDIA · audio$0.0001998/minute of audio
  15. 15NVIDIA: Parakeet TDT 0.6B v3NVIDIA · audio$0.0015/minute of audio
  16. 16OpenAI: GPT TranscribeOpenAI · audio$0.0045/minute of audio
  17. 17OpenAI: GPT-4o Mini TranscribeOpenAI · audio$1.25/M audio tokens · $5/M text tokens
  18. 18OpenAI: GPT-4o TranscribeOpenAI · audio$2.5/M audio tokens · $10/M text tokens
  19. 19OpenAI: Whisper 1OpenAI · audio$0.006/minute of audio
  20. 20OpenAI: Whisper Large V3OpenAI · audio$0.00185/minute of audio
  21. 21OpenAI: Whisper Large V3 TurboOpenAI · audio$0.0006667/minute of audio
  22. 22Qwen: Qwen3 ASR 0.6BQwen · audio$0.0001998/minute of audio
  23. 23Qwen: Qwen3 ASR 1.7BQwen · audio$0.00045/minute of audio
  24. 24SpaceXAI: Grok STT 1.0xAI · audio$0.001667/minute of audio

FAQ

How do I transcribe audio to text with Hormouz?
Use the audio/transcriptions endpoint with one of the 24 speech-to-text models listed here, such as AssemblyAI: Universal-3.5 Pro. One API key covers all of them.

Updated 2026-10-13