Update README with dataset statistics

#2
by dhakar - opened
Files changed (1) hide show
  1. README.md +109 -0
README.md CHANGED
@@ -133,3 +133,112 @@ The model demonstrates promising performance for Tibetan speech recognition, par
133
 
134
  ---
135
  *Report generated on July 14, 2025*
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
133
 
134
  ---
135
  *Report generated on July 14, 2025*
136
+
137
+
138
+ ---
139
+
140
+ # Dataset Statistics
141
+
142
+ ## Configuration: `default`
143
+
144
+ ### Split: `train`
145
+
146
+ **Total Rows**: 893
147
+
148
+ #### `audio_duration`
149
+
150
+ - **Type**: numerical
151
+ - **Data Type**: `float64`
152
+ - **Sum**: 3,870.19
153
+ - **Average**: 4.33
154
+
155
+ #### `strata`
156
+
157
+ - **Type**: categorical
158
+ - **Data Type**: `object`
159
+ - **Unique Values**: 8
160
+
161
+ **Value Distribution:**
162
+
163
+ | Value | Count | Percentage |
164
+ |-------|-------|------------|
165
+ | `70-80__medium__Prayer` | 125 | 14.00% |
166
+ | `70-80__short__Teaching` | 123 | 13.77% |
167
+ | `70-80__long__Teaching` | 123 | 13.77% |
168
+ | `70-80__long__Prayer` | 121 | 13.55% |
169
+ | `70-80__medium__Teaching` | 118 | 13.21% |
170
+ | `70-80__long__Q&A` | 108 | 12.09% |
171
+ | `80-90__long__Practice` | 108 | 12.09% |
172
+ | `70-80__medium__Q&A` | 67 | 7.50% |
173
+
174
+ #### `age_group`
175
+
176
+ - **Type**: categorical
177
+ - **Data Type**: `object`
178
+ - **Unique Values**: 2
179
+
180
+ **Value Distribution:**
181
+
182
+ | Value | Count | Percentage |
183
+ |-------|-------|------------|
184
+ | `70-80` | 785 | 87.91% |
185
+ | `80-90` | 108 | 12.09% |
186
+
187
+ #### `duration_category`
188
+
189
+ - **Type**: categorical
190
+ - **Data Type**: `object`
191
+ - **Unique Values**: 3
192
+
193
+ **Value Distribution:**
194
+
195
+ | Value | Count | Percentage |
196
+ |-------|-------|------------|
197
+ | `long` | 460 | 51.51% |
198
+ | `medium` | 310 | 34.71% |
199
+ | `short` | 123 | 13.77% |
200
+
201
+ #### `content_type`
202
+
203
+ - **Type**: categorical
204
+ - **Data Type**: `object`
205
+ - **Unique Values**: 4
206
+
207
+ **Value Distribution:**
208
+
209
+ | Value | Count | Percentage |
210
+ |-------|-------|------------|
211
+ | `Teaching` | 364 | 40.76% |
212
+ | `Prayer` | 246 | 27.55% |
213
+ | `Q&A` | 175 | 19.60% |
214
+ | `Practice` | 108 | 12.09% |
215
+
216
+ #### `base_cer`
217
+
218
+ - **Type**: numerical
219
+ - **Data Type**: `float64`
220
+ - **Sum**: 247.09
221
+ - **Average**: 0.28
222
+
223
+ #### `cer_5000`
224
+
225
+ - **Type**: numerical
226
+ - **Data Type**: `float64`
227
+ - **Sum**: 244.76
228
+ - **Average**: 0.27
229
+
230
+ #### `cer_10000`
231
+
232
+ - **Type**: numerical
233
+ - **Data Type**: `float64`
234
+ - **Sum**: 208.71
235
+ - **Average**: 0.23
236
+
237
+ #### `cer_19000`
238
+
239
+ - **Type**: numerical
240
+ - **Data Type**: `float64`
241
+ - **Sum**: 204.79
242
+ - **Average**: 0.23
243
+
244
+ ---