Upload sanitized corpus and readme

README.md

@@ -141,11 +141,18 @@ The dataset consists of a train, dev and test splits, for both short and long se
 
 ### Example Usage
 
-To load a specific split, for example, the training split (short) do:
+To load the dataset:
 
 ```python
-  from datasets import load_dataset
-  data = load_dataset("projecte-aina/corts_valencianes_asr_a",split="clean_train_short")
+from datasets import load_dataset
+data = load_dataset("projecte-aina/corts_valencianes_asr_a",trust_remote_code=True)
+```
+
+To load the dataset with streaming enabled (recommended for large audio files):
+
+```python
+from datasets import load_dataset
+data = load_dataset("projecte-aina/corts_valencianes_asr_a",trust_remote_code=True, streaming=True)
 ```
 
 ## Dataset Creation

corpus/speech/clean_dev_long/clean_dev_long_1.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc2c45788d43164ca7daee7cbe25480c8eb93719ddae5c6d9f4ca1a2593c8b56
-size 97521932
+oid sha256:1b509017c151cbe81d3b0cdc29c981b95c5a37f5c0c09adbfc94d32c9fdf3904
+size 97521676

corpus/speech/clean_dev_short/clean_dev_short_1.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:394bee56c07f4bc56f597648ae32771d0da2978655a88a6350bc02fb03b96066
-size 926915620
+oid sha256:ca40f838e0c0bbf74af3cb6cdc680bfc5bfc96317efbfab9c322eddc070a4a80
+size 926909148

corpus/speech/clean_test_long/clean_test_long_1.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f702326446f44bdea6547be53553e5d3f635ee4743b9533c3998d2a249f8031e
-size 103122566
+oid sha256:485dcfc2f660a1ce0d2d7b4a384fc4d7b4eab57da86f50ece11e97be704155d1
+size 103122434

corpus/speech/clean_test_short/clean_test_short_1.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f6f21228422bdd3e39166d6652dadfd80bb2b7fdf87852326998c3cc4ed92bb5
-size 903345015
+oid sha256:32852c5fb573d384a374ad612e20d62708d6276b1f81c27c859c4dd5aa6d36a1
+size 903338225

corpus/speech/clean_train_long/clean_train_long_1.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c8036c099fa5bc694da0f91803d7d6d2e7895f8aed54c6435831573105da1cab
-size 3006882179
+oid sha256:30adc51394e96628bd4281adbf5cd0ad2c30f41c2f64c10d94fc252e9f40b5fa
+size 3006863416

corpus/speech/clean_train_long/clean_train_long_2.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be52b36d06abb7855674c1671059ad9e8b701ea8063e57229ac82e0b2efe2bd8
-size 1376822990
+oid sha256:35a6030016e40a4288527e5b64b0c7ec52b6370a5ffe854428da227e0ca0988a
+size 1376823250

corpus/speech/clean_train_short/clean_train_short_1.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:267123cec6d77c1d5efdfd383f2ba28b9227d853257bddd31658ce3db8b086b7
-size 3007843689
+oid sha256:ea081d717d2d3a8411acfe9cb34c1d222b8d567a48e4a57f6df27b2980f58a38
+size 3007846575

corpus/speech/clean_train_short/clean_train_short_10.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9cd79a17bf96370271bce945a11064eef3c1f21490a54863574bc8cc767c7deb
-size 3008715257
+oid sha256:dbd53027e166d7b73b03bec959c1fede39a57771067203565f18866738de44bd
+size 3008664894

corpus/speech/clean_train_short/clean_train_short_11.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9616f5281545055834ce017a75dfc21e8ca5102bf1610f5c147916f219e621a9
-size 3008007112
+oid sha256:8d70b20e15932b0d2648cda8c5d288f41cca39c8e30c421835866faeefbab760
+size 3007985307

corpus/speech/clean_train_short/clean_train_short_12.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bbfaf8a08ab9d8ee3af5b5142c8412757c77efee2ee3814c4333c31df998fa7c
-size 3007828533
+oid sha256:c4dc988b9587a9b332f78a838d26fabfef478d3ace193d6d69d0e677da60d808
+size 3007799575

corpus/speech/clean_train_short/clean_train_short_13.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:748303085c661af97fe857c59b3544f052b74604b524336c6117f1d7400b651f
-size 135822471
+oid sha256:a67bf602e4359c8dfc85f5dec617bdd8e3753e7817e5e09e9aacbb6b79c020a9
+size 135822859

corpus/speech/clean_train_short/clean_train_short_2.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bfd3955af50f8b841591d92e6502ffade4e115a83b08782be85ba12d0f5b674b
-size 3007726448
+oid sha256:d31d3e8721e8748a00b519127e4e2c661b381c6c8d335e78d03cde97aff6031b
+size 3007728257

corpus/speech/clean_train_short/clean_train_short_3.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb78d523f7be7194211e9ec9cf22d649ebf0ac88b94ffe5c7ba23163ae851214
-size 3006586570
+oid sha256:854dcc185fdd00ad7ab2bbc066d29711cdae9e528f288b91ca38b3beaa188568
+size 3006581601

corpus/speech/clean_train_short/clean_train_short_4.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f7b7ecf4d7dd1c1f30cf81e3b7d32d5d2e8cfb451fcb4b11abd7d3f80c471ed6
-size 3008276947
+oid sha256:5fdb09eb56af69f3b4309625e5fcecad5214979e6f41dcd271fc9c69175b1556
+size 3008271083

corpus/speech/clean_train_short/clean_train_short_5.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ddf2454b3ab935035e05c518c9c5cf4d4e8915237dd2f5462b5c9588c5e6e688
-size 3007502315
+oid sha256:c93c67a61781fcff9c6c28671e47cf8dbea63a08462590b78b5c3d7c9a77637d
+size 3007512700

corpus/speech/clean_train_short/clean_train_short_6.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:98c01ee707708d2b13032697f125e7d58181661f6bc97f7807e884d83fad0882
-size 3008010627
+oid sha256:c9fb53acf1c461a61c733cd91cdbfec985ed75f2ec4c8ee936fb3d9ce1d4071f
+size 3008004975

corpus/speech/clean_train_short/clean_train_short_7.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb0bbe31d97878345e598da60f0cf67b2b8ca01afc9a9ca111eb0a35fe67a0e9
-size 3008451667
+oid sha256:cd3aac58ae2df1f33a8ee42142cd9a8f336691b585ef433a7880c7d29514fc7c
+size 3008400831

corpus/speech/clean_train_short/clean_train_short_8.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a171fc0f6f4414dcb7e29c3e0bcc6a932a86e70332897c2d249ad669779a8f8a
-size 3007976563
+oid sha256:4cf735a936085446674cb461be7f06b5fd9e6435a00668693c69dfd8f4080b9e
+size 3007930495

corpus/speech/clean_train_short/clean_train_short_9.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:492ba8e03a571e33d93a8d1367f0f3a5c5ec402c02c7b9289697c12a20f8fe47
-size 3007911731
+oid sha256:d185a36bb08ce2f39525984a45701775fded1cecd7025f41ba316eefe835210a
+size 3007912091

corpus/speech/other_dev_long/other_dev_long_1.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c67934d352fabcb228a58a36b795814a56d12da1f6a7c4f3e752e5dd7f1ebf8d
-size 87918915
+oid sha256:46fc5be17de455331c1ea1f1d55aac6e0e75fcbaaae95a478aebdf07f7e83d08
+size 87918751

corpus/speech/other_dev_short/other_dev_short_1.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ec675783c1ff293af392a9cd8eb9586747f11ec91d1de44a1e53cbd3c617bfb
-size 967037526
+oid sha256:b5444974c12183c1f7c59de18b691cdf8556f3604bda8fde1284bafd429a9ca1
+size 967027327

corpus/speech/other_test_long/other_test_long_1.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:02632147f9e894345200e85238d35822a85e16d3baa3f51c4baeb1ae58d209e6
-size 76099344
+oid sha256:366701933791576d82d8ce52e2e2d65ad4fb61ae535ad31ae4c47f3b11725afc
+size 76099242

corpus/speech/other_test_short/other_test_short_1.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4f317cf31772b57b7b827d98c128f4f04363e5e966e1806556f44b0b43fec15
-size 1001044024
+oid sha256:2a2c08e1faf581dd189392a78cfad9c45138a433ceaec9dd03d89085ea0c56ab
+size 1001018232

corpus/speech/other_train_long/other_train_long_1.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54a6dc41753e211466e6ab62cf1d386c35e6da029055ede3f410024386479fd3
-size 3002652092
+oid sha256:259879eccb711569b3d0b258a5e510cacfe7d0eed3457cd9a32b31b011e5569f
+size 3002647348

corpus/speech/other_train_long/other_train_long_2.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a9bac72565ac77ad61cc90586e5fa4b4f53621ec6312e4f117f0be9f8295dd6
-size 1977334128
+oid sha256:0f5d4e5aed1eb2e31c389967fc6873a452e5d827c72200d07a65d992a76cfcc8
+size 1977324386

corpus/speech/other_train_short/other_train_short_1.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:25fb49e9d81fed0482a6210fa898f8a46aab0e5d5d15f10a0c6abd6c0781aa5e
-size 3006463760
+oid sha256:56134f617c08ca751b3dada9eac35a94ee42c267304148a1e99618f4dbc12049
+size 3006462047

corpus/speech/other_train_short/other_train_short_10.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36a9fab166da3bf3d08edea0226fe3a125a943b63adaf861623311547d5c6699
-size 3006993355
+oid sha256:5b593af15eaaf60d74759035d03a8a42a6b31751ee0ef91fa0f3a647442d7e7b
+size 3006914452

corpus/speech/other_train_short/other_train_short_11.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f71dc8ec7eaeaf8f59bc391d5230fcd4f9a38cb1c140bbb7d1c9cd08e7be03c
-size 3006907031
+oid sha256:766c8a041c0ee9a900eb1bd76ecaf7dbc825f3318e44eaec98a4b68ca4f31e05
+size 3006884647

corpus/speech/other_train_short/other_train_short_12.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af54d8856bf0db2092aaca1725db92952653ad9739a7365808decfb6a77a2709
-size 2022091207
+oid sha256:db0e4d4c300696e072c849acd83b907e352aea6572c7ce02aa12f13a903aafef
+size 2022095865

corpus/speech/other_train_short/other_train_short_2.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9fe6bdd50944e6cf0c78ac110f82dea93db54d78c3b96765c5cc9a6d42c2f72c
-size 3007161443
+oid sha256:d92d2fb608c27a075752226d0d0c0d5aa9c3915cd4e2d73d23fefb23ec0b77e5
+size 3007185275

corpus/speech/other_train_short/other_train_short_3.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90cb52b367aa110c0f8ec742f56dee86bd17075349f0ab256e7121c486952ca9
-size 3006385813
+oid sha256:1f013e6b8c455584626aaf6f411e3734ef98f1cc2276b4ff5267b37e9468e8d8
+size 3006405506

corpus/speech/other_train_short/other_train_short_4.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7081ef7c344d4986c038eb955ffb5beff56b871611775dc562453457e0e448d
-size 3007162198
+oid sha256:47225cef25096f46ed04ebb46a2e3578f374e7bdc9628b45a36130fa83615357
+size 3007122383

corpus/speech/other_train_short/other_train_short_5.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ebb8da5bb223a56b613073fa3db837b281124777403c2556dcc8a94bc85d1822
-size 3006754798
+oid sha256:f25c59aab4fe6cd250e1c55c6cec2710d3e2144abe314c73b5a2fa5e69aaa97b
+size 3006677786

corpus/speech/other_train_short/other_train_short_6.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:52d2ed9730bd6ff13f68aed758e98f752391147c905a5ad6d5ef952a8d929c77
-size 3007083358
+oid sha256:f189daedb65922babc71b59fd71823398544673fe6bc0da6382adb59a79e0864
+size 3007040653

corpus/speech/other_train_short/other_train_short_7.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d3024e5579099596a4af79400a0e8d9929ed53ab3874a1a3be5f2eda941dc8ba
-size 3006772901
+oid sha256:fcd3d641ce02fcf6ddcfbea406d8ee3631c22bb824980690ed23ad90a1e3b6bb
+size 3006757412

corpus/speech/other_train_short/other_train_short_8.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:780004b71f791e1e4426b798126853aac164b8e9e5de5696b910b00396a3007d
-size 3007154847
+oid sha256:3933a039079b5709de95b4df7b035a5fb41972b15c8b8868bb2a5d564ebd74c1
+size 3007098880

corpus/speech/other_train_short/other_train_short_9.tar.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7307bc2b9dd2cb8e669af5066db5ce7791a3eca908a46738a41b9ad007e6bbd0
-size 3007040957
+oid sha256:76824dad2d39124b03367ad2ee486044e8c27a93bf3fec0f9880d3a74e4767b5
+size 3007069812

corts_valencianes_asr_a.py

@@ -107,7 +107,7 @@ class CortsValencianesASR(datasets.GeneratorBasedBuilder):
         metadata_other_train_short=dl_manager.download_and_extract(_METADATA_OTHER_TRAIN_SHORT)
         metadata_other_test_short=dl_manager.download_and_extract(_METADATA_OTHER_TEST_SHORT)
         metadata_other_dev_short=dl_manager.download_and_extract(_METADATA_OTHER_DEV_SHORT)
-
+        
         tars_clean_train_short=dl_manager.download_and_extract(_TARS_CLEAN_TRAIN_SHORT)
         tars_clean_test_short=dl_manager.download_and_extract(_TARS_CLEAN_TEST_SHORT)
         tars_clean_dev_short=dl_manager.download_and_extract(_TARS_CLEAN_DEV_SHORT)
@@ -132,38 +132,40 @@ class CortsValencianesASR(datasets.GeneratorBasedBuilder):
         tars_other_train_long=dl_manager.download_and_extract(_TARS_OTHER_TRAIN_LONG)
         tars_other_test_long=dl_manager.download_and_extract(_TARS_OTHER_TEST_LONG)
         tars_other_dev_long=dl_manager.download_and_extract(_TARS_OTHER_DEV_LONG)
-       
+        
         hash_tar_files=defaultdict(dict)
         with open(tars_clean_train_short,'r') as f:
-            hash_tar_files['clean_train_short']=[path.replace('\n','') for path in f]
+            hash_tar_files['clean_train_short']=[path.strip() for path in f if path.strip()]
         with open(tars_clean_test_short,'r') as f:
-            hash_tar_files['clean_test_short']=[path.replace('\n','') for path in f]
+            hash_tar_files['clean_test_short']=[path.strip() for path in f if path.strip()]
         with open(tars_clean_dev_short,'r') as f:
-            hash_tar_files['clean_dev_short']=[path.replace('\n','') for path in f]
+            hash_tar_files['clean_dev_short']=[path.strip() for path in f if path.strip()]
             
         with open(tars_other_train_short,'r') as f:
-            hash_tar_files['other_train_short']=[path.replace('\n','') for path in f]
+            hash_tar_files['other_train_short']=[path.strip() for path in f if path.strip()]
         with open(tars_other_test_short,'r') as f:
-            hash_tar_files['other_test_short']=[path.replace('\n','') for path in f]
+            hash_tar_files['other_test_short']=[path.strip() for path in f if path.strip()]
         with open(tars_other_dev_short,'r') as f:
-            hash_tar_files['other_dev_short']=[path.replace('\n','') for path in f]
+            hash_tar_files['other_dev_short']=[path.strip() for path in f if path.strip()]
             
 
         with open(tars_clean_train_long,'r') as f:
-            hash_tar_files['clean_train_long']=[path.replace('\n','') for path in f]
+            hash_tar_files['clean_train_long']=[path.strip() for path in f if path.strip()]
         with open(tars_clean_test_long,'r') as f:
-            hash_tar_files['clean_test_long']=[path.replace('\n','') for path in f]
+            hash_tar_files['clean_test_long']=[path.strip() for path in f if path.strip()]
         with open(tars_clean_dev_long,'r') as f:
-            hash_tar_files['clean_dev_long']=[path.replace('\n','') for path in f]
+            hash_tar_files['clean_dev_long']=[path.strip() for path in f if path.strip()]
             
         with open(tars_other_train_long,'r') as f:
-            hash_tar_files['other_train_long']=[path.replace('\n','') for path in f]
+            hash_tar_files['other_train_long']=[path.strip() for path in f if path.strip()]
         with open(tars_other_test_long,'r') as f:
-            hash_tar_files['other_test_long']=[path.replace('\n','') for path in f]
+            hash_tar_files['other_test_long']=[path.strip() for path in f if path.strip()]
         with open(tars_other_dev_long,'r') as f:
-            hash_tar_files['other_dev_long']=[path.replace('\n','') for path in f]
+            hash_tar_files['other_dev_long']=[path.strip() for path in f if path.strip()]
+        
             
-        hash_meta_paths={"clean_train_short":metadata_clean_train_short,
+        hash_meta_paths={
+        "clean_train_short":metadata_clean_train_short,
         "clean_test_short":metadata_clean_test_short,
         "clean_dev_short":metadata_clean_dev_short,
         "other_train_short":metadata_other_train_short,
@@ -175,7 +177,7 @@ class CortsValencianesASR(datasets.GeneratorBasedBuilder):
         "other_train_long":metadata_other_train_long,
         "other_test_long":metadata_other_test_long,
         "other_dev_long":metadata_other_dev_long}
-
+        
         audio_paths = dl_manager.download(hash_tar_files)
         
         splits=["clean_train_short","clean_test_short","clean_dev_short","other_train_short","other_test_short","other_dev_short","clean_train_long","clean_test_long","clean_dev_long","other_train_long","other_test_long","other_dev_long"]