Transcribe and translate audio with OpenAI's open Whisper model, handle long recordings, and measure accuracy with word error rate.
Whisper is an open speech recognition model that runs on your own machine. This course uses the Whisper tiny model from the hub to transcribe and translate audio, then covers the practical parts: long recordings, timestamps, choosing a model size, and measuring accuracy honestly.
Install with pip install transformers torch jiwer, plus ffmpeg for reading audio files.
4 lessons · 56 min
From sound waves to spectrograms to text.
Run Whisper from Python on a recording.
Chunking, timestamps for subtitles, and preparing audio.
Compare transcripts to a reference and read the result.
No ratings yet.
Sign in to leave a rating. Sign in
Use @handle to bring somebody in, and #tag to file it.
Nothing here yet. Ask a question, or say how you used it.
Sign in to join the discussion. Sign in
2 min read
2 min read
2 min read
2 min read
Classify images with a pretrained ResNet-50 in ONNX Runtime, then adapt a pretrained network to your own categories.
Trend, seasonality, honest baselines and time-based testing, using NASA temperatures and bike rentals from the hub.
Where the model sits in the workflow, who checks it, and what happens to the exceptions.