Hugging Face's logo

third-intelligence
/
RubatoCodec

Audio-to-Audio
Safetensors
audio
speech
neural-audio-codec
speech-tokenizer
variable-frame-rate
streaming
text-to-speech
Model card Files
xet
 Community
RubatoCodec
File size: 1,583 Bytes
2b291ec
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
{
  "semantic_model": "facebook/w2v-bert-2.0",
  "semantic_layer_index": 16,
  "sr": 16000,
  "output_sr": 44100,
  "codec_encoder": {
    "ngf": 48,
    "hidden_dim": 1024,
    "up_ratios": [
      2,
      2,
      4,
      4,
      5
    ],
    "dilations": [
      1,
      3,
      9
    ],
    "activation": "snake_beta"
  },
  "codec_decoder": {
    "hidden_dim": 1024,
    "activation": "snake_beta",
    "conv_padding_mode": "causal_full",
    "resnet_kernel_size": 1,
    "embed_kernel_size": 3,
    "depth": 12,
    "heads": 16,
    "pos_emb_dim": 64,
    "vq_dim": 2048,
    "attn_causal": true,
    "mlp_expansion": 4,
    "upsampler_factors": [
      3
    ],
    "upsampler_kernel_sizes": [
      9
    ],
    "output_n_fft": 1176,
    "output_hop_length": 294,
    "quantizer": {
      "mode": "rvq",
      "num_quantizers": 16,
      "codebook_size": 4096,
      "fsq_levels": [
        8,
        8,
        8,
        8
      ],
      "cb1_type": "fsq",
      "residual_type": "vq",
      "vq_codebook_dim": 8
    }
  },
  "firing": {
    "enabled": true,
    "mode": "adaptive",
    "target_ratio": 0.25,
    "max_gap": 64,
    "decision": "clock",
    "hnet_ms_offsets": [
      [
        1,
        0
      ],
      [
        1,
        1
      ],
      [
        2,
        1
      ]
    ],
    "head_in_budget": true,
    "aggregate": "mean",
    "signals": [
      "hnet_ms",
      "flux"
    ]
  },
  "inference": {
    "attention_window_seconds": 6.0,
    "decode_window_frames": 300,
    "decode_stride_frames": 100,
    "decode_crossfade_ms": 10.0
  }
}