All tools

aiquick/whisper-large-v3

audio-to-text

faster-whisper large-v3 on the GPU worker. Unlike the Parakeet tool it detects the spoken language, reports its confidence, and returns per-word timings — which is what you want for subtitles or for lining a transcript up against the audio.

Inference/gpu/transcribe

Input

Audio file *Up to 6 MB. WAV, MP3, M4A, FLAC, or OGG.
LanguageOptional ISO code, e.g. en. Leave blank and the model detects it.

Result

Idle
Fill in the input and run the tool to see results here.