Parakeet TDT 0.6b v2 (Transformers.js v4 Compatible)

Required files for Transformers.js

• config.json (with model_type: "nemo-conformer-tdt" and transformers.js_config.transducer)
• tokenizer.json
• tokenizer_config.json
• special_tokens_map.json
• preprocessor_config.json

• onnx/encoder_model.onnx
• onnx/decoder_model_merged.onnx
• optional dtype variants: *_fp16.onnx, *_int8.onnx

Notes

• onnx/encoder-model.onnx.data is the external data file referenced by onnx/encoder_model.onnx.
• vocab.txt is kept for provenance/conversion, but runtime uses tokenizer.json + tokenizer_config.json.
• nemo128.onnx at repo root is optional and currently not used by the current Transformers.js Nemo feature extractor path (JS mel extraction via preprocessor_config.json). Keeping it at root is fine.

Compatibility check

• AutoConfig.from_pretrained(...)
• AutoTokenizer.from_pretrained(...)
• AutoProcessor.from_pretrained(...)
• NemoConformerForTDT.from_pretrained(...)
• pipeline('automatic-speech-recognition', ...)

Provenance

• ATTRIBUTION.md
• provenance/manifest.json