Update sw_tz: +287.7h to 1,002.9h (Zanzibar v5 + new prep); add parliament_zanzibar_v5 paths to YAML

README.md

@@ -766,8 +766,11 @@ configs:
     path:
     - data/sw_tz/train-*.parquet
     - data/sw_tz/tz_sw_znz-train-*.parquet
+    - data/sw_tz/parliament_zanzibar_v5-train-*.parquet
   - split: test
-    path: data/sw_tz/test-*.parquet
+    path:
+    - data/sw_tz/test-*.parquet
+    - data/sw_tz/parliament_zanzibar_v5-test-*.parquet
 - config_name: ta_in
   data_files:
   - split: train
@@ -891,7 +894,7 @@ A multilingual ASR dataset containing 53k hours of transcribed speech across 113
 
 | Metric | Value |
 | :--- | :--- |
-| Total hours | 54,721 |
+| Total hours | 55,008 |
 | Distinct languages | 88 |
 | Language variations | 127 |
 | Languages ≥ 1,000 h | 23 |
@@ -1066,7 +1069,7 @@ for row in ds:
 | `st_za` | South Africa | Sesotho | 0.4 | 2.67 |
 | `sv_ax` | Åland Islands | Swedish | 66.0 | 2.73 |
 | `sw_ke` | Kenya | Swahili | 257.4 | 2.93 |
-| `sw_tz` | Tanzania | Swahili | 715.2 | 2.56 |
+| `sw_tz` | Tanzania | Swahili | 1,002.9 | 2.56 |
 | `ta_in` | India | Tamil | 36.1 | 3.06 |
 | `ta_lk` | Sri Lanka | Tamil | 204.0 | 2.80 |
 | `te_in` | India | Telugu | 77.0 | 3.13 |
@@ -1085,7 +1088,7 @@ for row in ds:
 | `yue_hk` | Hong Kong | Cantonese | 1,943.5 | 3.02 |
 | `zh_tw` | Taiwan | Mandarin | 1,482.1 | 2.47 |
 | `zu_za` | South Africa | Zulu | 19.0 | 2.71 |
-| **Total** | — | — | **54,720.6** | **2.83** |
+| **Total** | — | — | **55,008.3** | **2.83** |
 
 ⚠️ Five configs (`kh_km` Khmer, `la_lo` Lao, `mm_my` Burmese, `vn_vi` Vietnamese, `cn_ug` Uyghur, sourced from Radio Free Asia include only metadata, transcripts, and source URLs. Audio must be downloaded separately from `source_url`.