Transcribe audio with Whisper-Turbo

Whisper-Turbo balances quality and speed; Moonshine-tiny is a good edge default if you are bandwidth-constrained. Passing catalog-id supplies the decoding pipeline, audio window and Whisper checkpoint shape from the catalog entry, and applies its license tier. The three paths point at files already on the node.

tenzro transcribe catalog

tenzro transcribe load \
  --model asr \
  --encoder-path /models/whisper-turbo/encoder_model.onnx \
  --decoder-path /models/whisper-turbo/decoder_model_merged.onnx \
  --tokenizer-path /models/whisper-turbo/tokenizer.json \
  --catalog-id whisper-large-v3-turbo

The CLI handles mel-spectrogram preprocessing and BPE detokenization for you.

tenzro transcribe run --model asr --audio audio.wav

If you know the language ahead of time, pass it to avoid the language-ID step.

tenzro transcribe run \
  --model asr \
  --audio interview.wav \
  --language en \
  --timestamps

The RPC returns a transcript string, plus per-segment timestamps when timestamps is set.

curl -s https://rpc.tenzro.xyz -H 'content-type: application/json' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tenzro_transcribe","params":{"model_id":"asr","audio_base64":"…","language":"en","timestamps":true}}'