Instructions to use FiditeNemini/CallMe with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use FiditeNemini/CallMe with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download FiditeNemini/CallMe --local-dir CallMe
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Add CallMe voice stack: Moonshine Small Streaming (stt/), Gemma-4-E2B MLX 4-bit (llm/)
785d121 verified Download stt/streaming_config.json from FiditeNemini/CallMe: direct link, hf CLI and curl.
- Browser
- Download file 512 Bytes
-
https://huggingface.co/FiditeNemini/CallMe/resolve/main/stt/streaming_config.json
- Command line
-
hf download hf://FiditeNemini/CallMe/stt/streaming_config.json
-
curl -L -o streaming_config.json https://huggingface.co/FiditeNemini/CallMe/resolve/main/stt/streaming_config.json
512 Bytes
| { | |
| "encoder_dim": 620, | |
| "decoder_dim": 512, | |
| "depth": 10, | |
| "nheads": 8, | |
| "head_dim": 64, | |
| "vocab_size": 32768, | |
| "bos_id": 1, | |
| "eos_id": 2, | |
| "frame_len": 80, | |
| "total_lookahead": 16, | |
| "d_model_frontend": 620, | |
| "c1": 1240, | |
| "c2": 620, | |
| "frontend_state_shapes": { | |
| "sample_buffer": [ | |
| 1, | |
| 79 | |
| ], | |
| "sample_len": [ | |
| 1 | |
| ], | |
| "conv1_buffer": [ | |
| 1, | |
| 620, | |
| 4 | |
| ], | |
| "conv2_buffer": [ | |
| 1, | |
| 1240, | |
| 4 | |
| ], | |
| "frame_count": [ | |
| 1 | |
| ] | |
| } | |
| } |