{
  "architectures": [
    "NemotronStreamingRNNTForSpeechRecognition"
  ],
  "model_type": "nemotron_streaming_rnnt_mlx",
  "library_name": "mlx",
  "sample_rate": 16000,
  "vocab_size": 13087,
  "target": "nemo.collections.asr.models.rnnt_bpe_models_prompt.EncDecRNNTBPEModelWithPrompt",
  "preprocessor": {
    "sample_rate": 16000,
    "normalize": "NA",
    "window_size": 0.025,
    "window_stride": 0.01,
    "window": "hann",
    "features": 128,
    "n_fft": 512,
    "dither": 1e-05,
    "preemph": 0.97,
    "mag_power": 2.0
  },
  "encoder": {
    "feat_in": 128,
    "n_layers": 24,
    "d_model": 1024,
    "n_heads": 8,
    "ff_expansion_factor": 4,
    "subsampling_factor": 8,
    "self_attention_model": "rel_pos",
    "subsampling": "dw_striding",
    "conv_kernel_size": 9,
    "subsampling_conv_channels": 256,
    "pos_emb_max_len": 5000,
    "causal_downsampling": true,
    "use_bias": false,
    "xscaling": false,
    "conv_norm_type": "layer_norm"
  },
  "decoder": {
    "blank_as_pad": true,
    "vocab_size": 13087,
    "prednet": {
      "pred_hidden": 640,
      "pred_rnn_layers": 2
    }
  },
  "joint": {
    "num_classes": 13087,
    "jointnet": {
      "joint_hidden": 640,
      "activation": "relu",
      "encoder_hidden": 1024,
      "pred_hidden": 640
    }
  },
  "decoding": {
    "greedy": {
      "max_symbols": 10
    }
  },
  "prompt_kernel": {
    "num_prompts": 128,
    "hidden": 2048,
    "d_model": 1024
  },
  "streaming": {
    "chunk_ms": 320,
    "mel_frames": 32,
    "pre_cache_size": 9,
    "output_frames": 4,
    "attention_left_context": 56,
    "conv_cache_size": 8
  },
  "quantization": {
    "mode": "affine",
    "bits": 8,
    "group_size": 64
  }
}