Step 1: Upload an audio file or record with your microphone.
This demo supports audio files up to 30 mins long. You can transcribe longer files locally with this NeMo script.
Step 2: Choose the input and output language.
If input & output languages are the same, you can also toggle generating punctuation & capitalization and timestamps.
Step 3: Run the model.
🐤 Canary 1B Flash model | 🧑💻 NeMo Repository