Parakeet TDT 0.6B v3 for Core AI: 8-bit palettized encoder (0.65 GB), Mac and iPhone validated
56ae5ae verified | { | |
| "source": { | |
| "repo_id": "nvidia/parakeet-tdt-0.6b-v3", | |
| "revision": "541d1f99c6b0c3cd0b11a95167540bb8edefd82b", | |
| "files": [ | |
| "README.md", | |
| "config.json", | |
| "generation_config.json", | |
| "model.safetensors", | |
| "tokenizer.json", | |
| "tokenizer_config.json" | |
| ], | |
| "weights_sha256": "3a2026366188c8c68598edbbff92f8d11590a08e0ae2e6775544e7b07d6a5e11" | |
| }, | |
| "status": "exported", | |
| "components": { | |
| "frontend": { | |
| "status": "exported", | |
| "precision": "fp32", | |
| "inputs": [ | |
| { | |
| "name": "waveform", | |
| "shape": [ | |
| 1, | |
| 240000 | |
| ], | |
| "dtype": "float32" | |
| }, | |
| { | |
| "name": "valid_samples", | |
| "shape": [ | |
| 1 | |
| ], | |
| "dtype": "int32" | |
| } | |
| ], | |
| "outputs": [ | |
| "mel" | |
| ], | |
| "seconds": 0.7271555841434747 | |
| }, | |
| "predictor": { | |
| "status": "exported", | |
| "precision": "fp16-weights", | |
| "inputs": [ | |
| { | |
| "name": "token", | |
| "shape": [ | |
| 1, | |
| 1 | |
| ], | |
| "dtype": "int32" | |
| }, | |
| { | |
| "name": "hidden", | |
| "shape": [ | |
| 2, | |
| 1, | |
| 640 | |
| ], | |
| "dtype": "float32" | |
| }, | |
| { | |
| "name": "cell", | |
| "shape": [ | |
| 2, | |
| 1, | |
| 640 | |
| ], | |
| "dtype": "float32" | |
| } | |
| ], | |
| "outputs": [ | |
| "prediction", | |
| "next_hidden", | |
| "next_cell" | |
| ], | |
| "seconds": 0.42557008285075426 | |
| }, | |
| "joint": { | |
| "status": "exported", | |
| "precision": "fp16-weights", | |
| "inputs": [ | |
| { | |
| "name": "encoder", | |
| "shape": [ | |
| 1, | |
| 1, | |
| 640 | |
| ], | |
| "dtype": "float32" | |
| }, | |
| { | |
| "name": "predictor", | |
| "shape": [ | |
| 1, | |
| 1, | |
| 640 | |
| ], | |
| "dtype": "float32" | |
| } | |
| ], | |
| "outputs": [ | |
| "logits" | |
| ], | |
| "seconds": 0.14682633290067315 | |
| }, | |
| "encoder": { | |
| "status": "exported", | |
| "precision": "pal8-fp16", | |
| "inputs": [ | |
| { | |
| "name": "mel", | |
| "shape": [ | |
| 1, | |
| 1500, | |
| 128 | |
| ], | |
| "dtype": "float16" | |
| }, | |
| { | |
| "name": "valid_frames", | |
| "shape": [ | |
| 1 | |
| ], | |
| "dtype": "int32" | |
| } | |
| ], | |
| "outputs": [ | |
| "encoded" | |
| ], | |
| "seconds": 56.55972387478687 | |
| } | |
| }, | |
| "weights": { | |
| "encoder": "pal8", | |
| "predictor_joint": "fp16" | |
| }, | |
| "versions": { | |
| "torch": "2.11.0", | |
| "transformers": "5.17.0", | |
| "coreai-torch": "0.4.2", | |
| "coreai-core": "1.0.0b2" | |
| } | |
| } | |