Audio to PDF

Your files stay on this device

Live speech-to-text only works in browsers that implement SpeechRecognition, primarily Chromium-based browsers such as Chrome and Edge. On other browsers the tool falls back to pasting a transcript manually. No downloadable/offline speech model is bundled — transcription quality and availability depend entirely on the browser's built-in speech engine, and recognition may even be unavailable for languages your browser does not support. Speaker labels are user-entered placeholders, not automatic speaker diarization — the tool cannot tell you who spoke. Timestamps are approximate: the Web Speech API only exposes coarse result timing (`resultIndex` and final-result events), so per-utterance times should be treated as estimates. Transcription is not fully local by default: the browser may process audio through its own speech-recognition service. Conversion of the transcript to PDF happens entirely on-device and nothing is uploaded by this app.