Masterx/sherpa-onnx-nemotron-3.5-asr-streaming-0.6b-320ms-2026-06-11

Streaming Nemotron 3.5 ASR (0.6B, multilingual) — 320 ms — sherpa-onnx ONNX

sherpa-onnx streaming transducer export of nvidia/nemotron-3.5-asr-streaming-0.6b at the 320 ms cache-aware chunk (att_context_size = [56, 3]), produced with sherpa-onnx's official scripts/nemo/nemotron-3.5-asr-streaming-0.6b/export_onnx.py.

Ships the full package:

• fp32: encoder.onnx + encoder.data, decoder.onnx, joiner.onnx
• int8: encoder.int8.onnx, decoder.int8.onnx, joiner.int8.onnx (ORT dynamic)
• fp16: encoder.fp16.onnx, decoder.fp16.onnx, joiner.fp16.onnx (onnxruntime float16, keep_io_types)
• tokens.txt

The multilingual prompt_index encoder input is exposed for per-stream language selection (or auto = the metadata auto_prompt_id). Sibling of the -1120ms export.

Made for WinSTT.