add model voice-filter

config.json

@@ -1,20 +1,8 @@
 {
-  "_name_or_path": "./model-bin/voice_enhancing",
   "architectures": [
-    "ASRVoiceFilter"
+    "VoiceFilter"
   ],
   "audio_max_lengh": 20,
-  "blank_token_id": 30,
-  "ctc_zero_infinity": true,
-  "d_dec": 640,
-  "d_enc": 256,
-  "d_inner": 0,
-  "dec_drop": 0.2,
-  "do_asr": false,
-  "do_enh": true,
-  "dropout": 0.2,
-  "emb_drop": 0.2,
-  "enc_drop": 0.2,
   "enh_args": {
     "batch_size": 8,
     "batch_type": "folded",
@@ -104,25 +92,8 @@
     "xvector_emb_dim": 512
   },
   "enh_chunk_size": 5,
-  "freq_kn": 3,
-  "freq_std": 2,
-  "ignore_token_id": -1,
-  "label_smooth": 0.1,
-  "loss_reduction": "mean",
-  "model_type": "asr_voicefilter",
-  "n_classes_ctc": 33,
-  "n_classes_s2s": 5003,
-  "n_dec": 1,
-  "n_enc": 16,
-  "n_fft": 512,
-  "n_head": 4,
-  "n_kernel": 25,
-  "n_mels": 80,
+  "model_type": "voicefilter",
   "sample_rate": 16000,
-  "shared_emb": true,
-  "teacher_force": 0.7,
-  "time_ds": 1,
   "torch_dtype": "float32",
-  "transformers_version": "4.24.0",
-  "use_cnn": true
+  "transformers_version": "4.25.1"
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e0d535c3363e6131ba47f2ac832a11d12a698fb01c10aad72fed5193ac43284
-size 197748277
+oid sha256:493efe8f21ce5d9ca8aff7b9b9700f03883f64ba453ca2be4cf7c389445fe151
+size 217276167

xvector_sincnet.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:4bcec986de13da7af7ac88736572692359950df63669989c4f78b294934c9089
-size 96383626