fix the bug of the speed perturb

examples/speech_to_text/prep_librispeech_data.py

@@ -74,7 +74,7 @@ def process(args):
         print("Generating manifest...")
         print("Generating manifest...")
         for split in SPLITS:
         for split in SPLITS:
             manifest = {c: [] for c in MANIFEST_COLUMNS}
             manifest = {c: [] for c in MANIFEST_COLUMNS}
-            dataset = LIBRISPEECH(out_root.as_posix(), url=split)
+            dataset = LIBRISPEECH(data_root.as_posix(), url=split)
             for wav, sample_rate, utt, spk_id, chapter_no, utt_no in tqdm(dataset):
             for wav, sample_rate, utt, spk_id, chapter_no, utt_no in tqdm(dataset):
                 sample_id = f"{spk_id}-{chapter_no}-{utt_no}"
                 sample_id = f"{spk_id}-{chapter_no}-{utt_no}"
                 manifest["id"].append(sample_id)
                 manifest["id"].append(sample_id)

examples/speech_to_text/prep_mustc_data.py

@@ -243,7 +243,7 @@ def process(args):
                 manifest = {c: [] for c in MANIFEST_COLUMNS}
                 manifest = {c: [] for c in MANIFEST_COLUMNS}
                 if args.task == "st" and args.add_src:
                 if args.task == "st" and args.add_src:
                     manifest["src_text"] = []
                     manifest["src_text"] = []
-                dataset = MUSTC(args.data_root, lang, split)
+                dataset = MUSTC(args.data_root, lang, split, args.speed_perturb)
                 for idx in range(len(dataset)):
                 for idx in range(len(dataset)):
                     items = dataset.get_fast(idx)
                     items = dataset.get_fast(idx)
                     for item in items:
                     for item in items: