Instructions to use konradjr007/nemotron-3-diarization-litert with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LiteRT
How to use konradjr007/nemotron-3-diarization-litert with LiteRT:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Nemotron-3 Diarization (LiteRT)
On-device streaming speaker diarization system (who spoke when) executing via LiteRT raw graphs.
Pipeline Components
| Component | File Path | Format | Role |
|---|---|---|---|
| Frontend | gpu/nemotron3_diar_frontend.tflite | LiteRT Graph | Mel spectrogram extraction |
| Streaming Encoder | gpu/nemotron3_diar_encoder_low_latency_fp16.tflite | LiteRT Graph | Low-latency streaming speaker embeddings |
| Offline Encoder | gpu/nemotron3_diar_encoder_offline_fp16.tflite | LiteRT Graph | High-accuracy file-level encoder |
| Mel Filterbank | gpu/frontend_mel128_257.bin | Binary Table | 128x257 Mel filterbank weights |
| Hann Window | gpu/hann400.bin | Binary Table | 400-point Hann window coefficients |
| Silence Embedding | gpu/silence_embeds.bin | Binary Table | Reference silence feature embeddings |
- Downloads last month
- 84
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐ Ask for provider support