Audio to Text

Transcribe English audio locally with Whisper.

Choose a local file (max 120 MiB)The English Whisper model and ONNX runtime download from Hugging Face on first use; model weight size depends on the selected quantization. The audio file stays local. Transcription may miss words, especially with accents, noise, overlapping speakers, or specialist terms.Your file is processed on this device. Processing models may download separately.