# Pre-recorded Speech-to-Text API

AssemblyAI's Pre-recorded Speech-to-Text API creates accurate transcripts from recorded audio and video.

Canonical URL: https://www.assemblyai.com/products/speech-to-text

## Summary

Use this API when the audio is already recorded. It supports high-accuracy transcription, language detection, word-level timestamps, speaker labels, custom spelling, prompting, keyterms, medical terminology, and transcript post-processing features.

## Core Models

- Universal-3.5 Pro: highest-accuracy pre-recorded model, recommended default for supported languages and difficult real-world audio.
- Universal-2: broad language coverage and legacy feature support.

## Common Use Cases

- Meeting transcription and notes.
- Contact center recordings.
- Medical documentation.
- Media transcription and captions.
- Sales call intelligence.
- Dataset preparation and searchable archives.

## Related References

- https://www.assemblyai.com/docs/pre-recorded-audio/getting-started/transcribe-an-audio-file.md
- https://www.assemblyai.com/docs/pre-recorded-audio/select-the-speech-model.md
- https://www.assemblyai.com/docs/pre-recorded-audio/universal-3-pro.md
- https://www.assemblyai.com/docs/pre-recorded-audio/keyterms-prompting.md
- https://www.assemblyai.com/docs/pre-recorded-audio/medical-mode.md
- https://www.assemblyai.com/pricing.md