Snap
Browse files- data.pkl → dataset.pkl +2 -2
- vps_clustering_benchmark.py +2 -9
data.pkl → dataset.pkl
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7be56f2d87b72d421e97b8158176eebf3317e0c4ac697e4a9fa671ad1eec949a
|
| 3 |
+
size 198842
|
vps_clustering_benchmark.py
CHANGED
|
@@ -4,14 +4,7 @@ import numpy as np
|
|
| 4 |
|
| 5 |
logger = datasets.logging.get_logger(__name__)
|
| 6 |
|
| 7 |
-
|
| 8 |
-
This dataset consists of a small sample of audio clips with annotated
|
| 9 |
-
speaker identities, their age and gender, diarization-based speech segments,
|
| 10 |
-
and transcription. The dataset is intended for the benchmarking of VBI-core.
|
| 11 |
-
"""
|
| 12 |
-
|
| 13 |
-
_HF_REPO_PATH = "https://huggingface.co/datasets/conversy/vps_clustering_benchmark/resolve/main/data.pkl"
|
| 14 |
-
|
| 15 |
|
| 16 |
class VPClusteringBenchmarkConfig(datasets.BuilderConfig):
|
| 17 |
"""BuilderConfig for Conversy Benchmark."""
|
|
@@ -46,7 +39,7 @@ class VPClusteringBenchmark(datasets.GeneratorBasedBuilder):
|
|
| 46 |
features=["segment_id", "filename", "speaker", "duration", "vp",
|
| 47 |
"start", "end", "readable_start", "readable_end",
|
| 48 |
"segment_clean"],
|
| 49 |
-
data_url=
|
| 50 |
nb_data_shards=1)
|
| 51 |
]
|
| 52 |
|
|
|
|
| 4 |
|
| 5 |
logger = datasets.logging.get_logger(__name__)
|
| 6 |
|
| 7 |
+
_DATA_PATH = "https://huggingface.co/datasets/conversy/vps_clustering_benchmark/resolve/main/dataset.pkl"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
|
| 9 |
class VPClusteringBenchmarkConfig(datasets.BuilderConfig):
|
| 10 |
"""BuilderConfig for Conversy Benchmark."""
|
|
|
|
| 39 |
features=["segment_id", "filename", "speaker", "duration", "vp",
|
| 40 |
"start", "end", "readable_start", "readable_end",
|
| 41 |
"segment_clean"],
|
| 42 |
+
data_url=_DATA_PATH,
|
| 43 |
nb_data_shards=1)
|
| 44 |
]
|
| 45 |
|