config.json

{
  "format": "naptrap-google-speech-embedding-v1",
  "embedding_size": 96,
  "parameters": 329448,
  "preprocessing": {
    "sample_rate": 16000,
    "leading_zeros": 11120,
    "frames": "floor(samples / 1280)",
    "stft": "400-sample Hann window, hop 160, 512-point FFT, no centering, magnitude",
    "mel": "32 bins, natural log(mel + 0.001)"
  },
  "source": {
    "publisher": "Google",
    "model": "google/speech_embedding/1 (TensorFlow1)",
    "url": "https://www.kaggle.com/models/google/speech-embedding/TensorFlow1/speech-embedding/1",
    "license": "Apache-2.0",
    "archive_sha256": "0f626c42009c1eb916401b7b2800229a211e43100115bc591b9afbf1b60da27f"
  },
  "modifications": [
    "TensorFlow graph re-implemented with PyTorch operations",
    "HWIO convolution kernels transposed to OIHW",
    "the original inference-time spatial mean subtraction applied to the first kernel",
    "original frontend constants and all trained parameters/statistics retained; no training"
  ],
  "sha256": {
    "model.safetensors": "80a10015f4eb8ff9ba1aa10b56b1d6e27ef9f9f3419ebbae56fb43e9afce37be"
  }
}