Upload 6 files

speaker_similarity/pyannote/pyannote_diarization_config.yaml

@@ -0,0 +1,21 @@
+version: 3.1.0
+
+pipeline:
+  name: pyannote.audio.pipelines.SpeakerDiarization
+  params:
+    clustering: AgglomerativeClustering
+    # embedding: pyannote/wespeaker-voxceleb-resnet34-LM  # if you want to use the HF model
+    embedding: /star-home/zhuhan/model/huggingface/pyannote/pyannote_model_wespeaker-voxceleb-resnet34-LM.bin  # if you want to use the local model
+    embedding_batch_size: 32
+    embedding_exclude_overlap: true
+    # segmentation: pyannote/segmentation-3.0  # if you want to use the HF model
+    segmentation: /star-home/zhuhan/model/huggingface/pyannote/pyannote_model_segmentation-3.0.bin  # if you want to use the local model
+    segmentation_batch_size: 32
+
+params:
+  clustering:
+    method: centroid
+    min_cluster_size: 12
+    threshold: 0.7045654963945799
+  segmentation:
+    min_duration_off: 0.0

speaker_similarity/pyannote/pyannote_model_segmentation-3.0.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da85c29829d4002daedd676e012936488234d9255e65e86dfab9bec6b1729298
+size 5905440

speaker_similarity/pyannote/pyannote_model_wespeaker-voxceleb-resnet34-LM.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:366edf44f4c80889a3eb7a9d7bdf02c4aede3127f7dd15e274dcdb826b143c56
+size 26645418

speaker_similarity/wavlm_large/hubconf.py

@@ -0,0 +1,78 @@
+# Copyright (c) Microsoft Corporation.
+# Licensed under the MIT License.
+
+# -*- coding: utf-8 -*- #
+"""*********************************************************************************************"""
+#   FileName     [ upstream/wavlm/hubconf.py ]
+#   Synopsis     [ the WavLM torch hubconf ]
+#   Author       [ Microsoft ]
+"""*********************************************************************************************"""
+
+
+import os
+
+from s3prl.util.download import _urls_to_filepaths
+
+from s3prl.upstream.wavlm.expert import UpstreamExpert as _UpstreamExpert
+
+
+def wavlm_local(ckpt, *args, **kwargs):
+    """
+    The model from local ckpt
+        ckpt (str): PATH
+    """
+    assert os.path.isfile(ckpt)
+    return _UpstreamExpert(ckpt, *args, **kwargs)
+
+
+def wavlm_url(ckpt, refresh=False, *args, **kwargs):
+    """
+    The model from google drive id
+        ckpt (str): URL
+        refresh (bool): whether to download ckpt/config again if existed
+    """
+    return wavlm_local(_urls_to_filepaths(ckpt, refresh=refresh), *args, **kwargs)
+
+
+def wavlm(refresh=False, *args, **kwargs):
+    """
+    The default model - Base-Plus
+        refresh (bool): whether to download ckpt/config again if existed
+    """
+    return wavlm_base_plus(refresh=refresh, *args, **kwargs)
+
+
+def wavlm_base(refresh=False, *args, **kwargs):
+    """
+    The Base model
+        refresh (bool): whether to download ckpt/config again if existed
+    """
+    kwargs[
+        "ckpt"
+    ] = "https://huggingface.co/s3prl/converted_ckpts/resolve/main/wavlm_base.pt"
+
+    return wavlm_url(refresh=refresh, *args, **kwargs)
+
+
+def wavlm_base_plus(refresh=False, *args, **kwargs):
+    """
+    The Base-Plus model
+        refresh (bool): whether to download ckpt/config again if existed
+    """
+    kwargs[
+        "ckpt"
+    ] = "https://huggingface.co/s3prl/converted_ckpts/resolve/main/wavlm_base_plus.pt"
+
+    return wavlm_url(refresh=refresh, *args, **kwargs)
+
+
+def wavlm_large(refresh=False, *args, **kwargs):
+    """
+    The Large model
+        refresh (bool): whether to download ckpt/config again if existed
+    """
+    kwargs[
+        "ckpt"
+    ] = "https://huggingface.co/s3prl/converted_ckpts/resolve/main/wavlm_large.pt"
+
+    return wavlm_url(refresh=refresh, *args, **kwargs)

speaker_similarity/wavlm_large/wavlm_large.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fb4b3c3e6aa567f0a997b30855859cb81528ee8078802af439f7b2da0bf100f
+size 1261965425

speaker_similarity/wavlm_large_finetune.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51f07e3b94d9e0262a6a675ef5a087be3dd09e8c62e9d886827f44f82fe7f94b
+size 1301926579