Skip to content

SOTA Open Source TTS

handle bad files

Lengyue6ac74ddea5b8
diff --git a/tools/vqgan/extract_vq.py b/tools/vqgan/extract_vq.py
index 419f48d..f6a3c9d 100644
--- a/tools/vqgan/extract_vq.py
+++ b/tools/vqgan/extract_vq.py
@@ -67,18 +67,28 @@ def get_model(
def process_batch(files: list[Path], model) -> float:
wavs = []
audio_lengths = []
+ new_files = []
max_length = total_time = 0
for file in files:
- wav, sr = torchaudio.load(file)
+ try:
+ wav, sr = torchaudio.load(file)
+ except Exception as e:
+ logger.error(f"Error reading {file}: {e}")
+ continue
+
if wav.shape[0] > 1:
wav = wav.mean(dim=0, keepdim=True)
wav = torchaudio.functional.resample(wav.cuda(), sr, model.sampling_rate)[0]
- wavs.append(wav)
total_time += len(wav) / model.sampling_rate
max_length = max(max_length, len(wav))
+
+ wavs.append(wav)
audio_lengths.append(len(wav))
+ new_files.append(file)
+
+ files = new_files
# Pad to max length
for i, wav in enumerate(wavs):
@@ -161,6 +171,8 @@ def main(
else:
files = list_files(folder, AUDIO_EXTENSIONS, recursive=True, sort=True)
+ print(f"Found {len(files)} files")
+ files = [Path(f) for f in files if not Path(f).with_suffix(".npy").exists()]
Random(42).shuffle(files)
total_files = len(files)