📄 update README.md

README.md

@@ -1,35 +1,61 @@
----
-dataset_info:
-  features:
-  - name: audio_id
-    dtype: string
-  - name: speaker
-    dtype: string
-  - name: start_seconds
-    dtype: float32
-  - name: end_seconds
-    dtype: float32
-  - name: start_ts
-    dtype: string
-  - name: end_ts
-    dtype: string
-  - name: transcript
-    dtype: string
-  - name: overlap
-    dtype: bool
-  - name: audio
-    dtype:
-      audio:
-        sampling_rate: 16000
-  splits:
-  - name: train
-    num_bytes: 431722063.0
-    num_examples: 953
-  download_size: 429777923
-  dataset_size: 431722063.0
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
----
+---
+license: cc-by-4.0
+tags:
+  - audio-diarization
+  - speaker-segmentation
+language:
+  - kk
+---
+
+# Diarizations Dataset
+
+Aggregated speaker segmentation outputs.
+
+## Videos
+
+### `ecx7ywj89m4`
+- Segments: **54**
+- Speakers: **SPEAKER_00, SPEAKER_01**
+- Duration: **2369.59s**
+#### Config
+- Trimmed: none
+- Diarize model: `pyannote/speaker-diarization`
+- ASR model: `akuzdeuov/whisper-base.kk` chunk `30s`, lang `kk`
+- Batch: `16`
+
+### `FWmr-zrGK_w`
+- Segments: **206**
+- Speakers: **SPEAKER_00, SPEAKER_01, SPEAKER_02**
+- Duration: **3071.74s**
+#### Config
+- Trimmed: none
+- Diarize model: `pyannote/speaker-diarization`
+- ASR model: `akuzdeuov/whisper-base.kk` chunk `30s`, lang `kk`
+- Batch: `16`
+
+### `FzeUEbA6j4E`
+- Segments: **535**
+- Speakers: **SPEAKER_00, SPEAKER_01, SPEAKER_02, SPEAKER_03**
+- Duration: **4425.93s**
+#### Config
+- Trimmed: none
+- Diarize model: `pyannote/speaker-diarization`
+- ASR model: `akuzdeuov/whisper-base.kk` chunk `30s`, lang `kk`
+- Batch: `16`
+
+### `WC_8foLpFbE`
+- Segments: **158**
+- Speakers: **SPEAKER_00, SPEAKER_01, SPEAKER_02**
+- Duration: **3605.37s**
+#### Config
+- Trimmed: none
+- Diarize model: `pyannote/speaker-diarization`
+- ASR model: `akuzdeuov/whisper-base.kk` chunk `30s`, lang `kk`
+- Batch: `16`
+
+## Usage
+```python
+from datasets import load_dataset
+ds = load_dataset('pushthetempo/diarizations')
+print(ds)
+```