Downloads · 30 days
9
8% of all-time downloads
goryodog/tokihisu-nemotron-3.5-asr-streaming-0.6b-webgpu-fp16
tokihisu-nemotron-3.5-asr-streaming-0.6b-webgpu-fp16 is a automatic speech recognition model from goryodog. Use it when you need speech turned into text. It is set up for transformers.js. The card lists the license as openmdw-1.1.
This repository is a custom three-session FP16 ONNX export of nvidia/nemotron-3.5-asr-streaming-0.6b for browser WebGPU use.
Downloads · 30 days
9
8% of all-time downloads
All-time downloads
115
Public
Repo size
1.3 GB
Likes
0
Public
Click a slice to open those files.
.onnx_data1.3 GB · 100%
From the Hugging Face model README
This repository is a custom three-session FP16 ONNX export of
nvidia/nemotron-3.5-asr-streaming-0.6b for browser WebGPU use.
f3d333391852ba876df169dcc9ba902d25b6ab0bja-JP3 subsampled framesfp16fp32onnx/encoder_fp16.onnx: log-Mel features and language prompt -> projected encoder statesonnx/predictor_fp16.onnx: one RNNT predictor step with explicit LSTM hidden/cell stateonnx/joint_fp16.onnx: one RNNT joint step -> vocabulary logitsmanifest.json: exact source revision, graph contracts, package versions, and SHA-256 hashesLarge tensors are stored as ONNX external data. Keep every .onnx file together with its
referenced *_data file.
Transformers.js does not currently include a native Nemotron3_5AsrForRNNT pipeline/model
implementation. This repository is therefore not loaded with a one-line
pipeline('automatic-speech-recognition', ...) call. Use the Transformers.js ONNX backend
(or ONNX Runtime Web directly) and implement the RNNT greedy loop across encoder, predictor,
and joint sessions.
The encoder graph is the cache-free, chunk-limited encoder path from the official Transformers implementation. For continuous recognition, feed a rolling/padded log-Mel window and retain the RNNT predictor state. It does not expose the official cache-aware encoder K/V and convolution caches in this version of the package.
Read manifest.json.contracts rather than hard-coding dimensions. The principal inputs are:
input_features, attention_mask, prompt_idsinput_ids, hidden_state, cell_stateencoder_frame, predictor_outputThe source processor uses 16 kHz audio and 128-bin log-Mel features. Copying the original
processor/tokenizer files into this repository does not by itself add a Nemotron processor class
to Transformers.js; the application must produce features matching processor_config.json.
The upstream model is distributed under Open Model Development and Weights License 1.1. Review the upstream model card and license terms before redistribution or deployment.