prepare dataset for fine-tuning whisper

create_nmsqacsv.py

@@ -0,0 +1,48 @@
+import wave, struct
+import numpy as np
+import datasets
+import transformers
+import os
+import pyarrow.parquet as pq
+from scipy.io.wavfile import write, read
+from datasets import Audio
+import IPython.display as ipd
+
+dataset = datasets.load_dataset("voidful/NMSQA_audio")
+train_set = dataset['train']
+dev_set = dataset['dev']
+train_dir, dev_dir = "./NMSQA-train-wav", "./NMSQA-dev-wav"
+os.makedirs(train_dir, exist_ok=True)
+os.makedirs(dev_dir, exist_ok=True)
+with open("NMSQA-train.csv", "w") as f:
+    f.writelines("path,text\n")
+with open("NMSQA-dev.csv", "w") as f:
+    f.writelines("path,text\n")
+
+for data in train_set:
+    try:
+        audio_data = data['content_segment_audio_path']['array']
+        audio = ipd.Audio(data=audio_data, autoplay=False, rate=24000)
+    except:
+        continue
+    gt_text = data['content_segment_normalized_text']
+    ids = data['id']
+    file_name = ids + '.wav'
+    with open(os.path.join(train_dir, file_name), "wb") as f:
+        f.write(audio.data)
+    with open("NMSQA-train.csv", "a") as f:
+        f.writelines(os.path.join(train_dir, file_name) + ',' + gt_text + "\n")
+
+for data in dev_set:
+    try:
+        audio_data = data['content_segment_audio_path']['array']
+        audio = ipd.Audio(data=audio_data, autoplay=False, rate=24000)
+    except:
+        continue
+    gt_text = data['content_segment_normalized_text']
+    ids = data['id']
+    file_name = ids + '.wav'
+    with open(os.path.join(dev_dir, file_name), "wb") as f:
+        f.write(audio.data)
+    with open("NMSQA-dev.csv", "a") as f:
+        f.writelines(os.path.join(dev_dir, file_name) + ',' + gt_text + "\n")