Update files from the datasets library (from 1.13.3)

Release notes: https://github.com/huggingface/datasets/releases/tag/1.13.3

README.md

@@ -529,6 +529,10 @@ A typical data point comprises the path to the audio file, called path and its s
 ```
 {
   'path': '/home/cahya/.cache/huggingface/datasets/downloads/extracted/4d9cf915efc21110199074da4d492566dee6097068b07a680f670fcec9176e62/su_id_female/wavs/suf_00297_00037352660.wav'
+  'audio': {'path': '/home/cahya/.cache/huggingface/datasets/downloads/extracted/4d9cf915efc21110199074da4d492566dee6097068b07a680f670fcec9176e62/su_id_female/wavs/suf_00297_00037352660.wav',
+	  'array': array([-0.00048828, -0.00018311, -0.00137329, ...,  0.00079346,
+			  0.00091553,  0.00085449], dtype=float32),
+	  'sampling_rate': 16000},
   'sentence': 'Panonton ting haruleng ningali Kelly Clarkson keur nyanyi di tipi',
 }
 ```
@@ -537,6 +541,8 @@ A typical data point comprises the path to the audio file, called path and its s
 
 path: The path to the audio file
 
+audio: A dictionary containing the path to the downloaded audio file, the decoded audio array, and the sampling rate. Note that when accessing the audio column: `dataset[0]["audio"]` the audio file is automatically decoded and resampled to `dataset.features["audio"].sampling_rate`. Decoding and resampling of a large number of audio files might take a significant amount of time. Thus it is important to first query the sample index before the `"audio"` column, *i.e.* `dataset[0]["audio"]` should **always** be preferred over `dataset["audio"][0]`.
+
 sentence: The sentence the user was prompted to speak
 
 ### Data Splits

openslr.py

@@ -537,6 +537,7 @@ class OpenSlr(datasets.GeneratorBasedBuilder):
         features = datasets.Features(
             {
                 "path": datasets.Value("string"),
+                "audio": datasets.features.Audio(sampling_rate=48_000),
                 "sentence": datasets.Value("string"),
             }
         )
@@ -591,7 +592,7 @@ class OpenSlr(datasets.GeneratorBasedBuilder):
                     path = str(path_to_data.resolve())
                     sentence = sentence_index[filename]
                     counter += 1
-                    yield counter, {"path": path, "sentence": sentence}
+                    yield counter, {"path": path, "audio": path, "sentence": sentence}
         else:
             for i, path_to_index in enumerate(path_to_indexs):
                 with open(path_to_index, encoding="utf-8") as f:
@@ -607,4 +608,4 @@ class OpenSlr(datasets.GeneratorBasedBuilder):
                         # set absolute path for audio file
                         path = os.path.join(path_to_datas[i], f"{filename}.wav")
                         counter += 1
-                        yield counter, {"path": path, "sentence": sentence}
+                        yield counter, {"path": path, "audio": path, "sentence": sentence}