Spaces:
Sleeping
Sleeping
Update inference_from_video.py
Browse files- inference_from_video.py +9 -18
inference_from_video.py
CHANGED
@@ -65,16 +65,16 @@ def parse_args():
|
|
65 |
help="How many test instances to evaluate.",
|
66 |
)
|
67 |
parser.add_argument(
|
68 |
-
"--sample_rate", type=int, default
|
69 |
-
help="Sample rate for audio output."
|
70 |
)
|
71 |
parser.add_argument(
|
72 |
"--save_dir", type=str, default="./outputs/tmp",
|
73 |
-
help="
|
74 |
)
|
75 |
parser.add_argument(
|
76 |
"--data_path", type=str, default="data/video_processed/video_gt_augment",
|
77 |
-
help="
|
78 |
)
|
79 |
|
80 |
args = parser.parse_args()
|
@@ -183,7 +183,7 @@ def main():
|
|
183 |
mel = vae.decode_first_stage(latents)
|
184 |
wave = vae.decode_to_waveform(mel)
|
185 |
|
186 |
-
all_outputs += [item
|
187 |
|
188 |
# Save #
|
189 |
exp_id = str(int(time.time()))
|
@@ -194,12 +194,7 @@ def main():
|
|
194 |
output_dir = "{}/{}_{}_steps_{}_guidance_{}_sampleRate_{}_augment".format(args.save_dir, exp_id, "_".join(args.model.split("/")[1:-1]), num_steps, guidance, sample_rate)
|
195 |
os.makedirs(output_dir, exist_ok=True)
|
196 |
for j, wav in enumerate(all_outputs):
|
197 |
-
|
198 |
-
try:
|
199 |
-
sf.write(file_path, wav, samplerate=sample_rate)
|
200 |
-
print(f"Saved {file_path}")
|
201 |
-
except Exception as e:
|
202 |
-
print(f"Error saving {file_path}: {e}")
|
203 |
|
204 |
else:
|
205 |
for i in range(num_samples):
|
@@ -213,12 +208,8 @@ def main():
|
|
213 |
ranked_wavs_for_text = [wavs_for_text[r] for r in rank]
|
214 |
|
215 |
for i, wav in enumerate(ranked_wavs_for_text):
|
216 |
-
|
217 |
-
|
218 |
-
sf.write(file_path, wav, samplerate=sample_rate)
|
219 |
-
print(f"Saved {file_path}")
|
220 |
-
except Exception as e:
|
221 |
-
print(f"Error saving {file_path}: {e}")
|
222 |
|
223 |
if __name__ == "__main__":
|
224 |
-
main()
|
|
|
65 |
help="How many test instances to evaluate.",
|
66 |
)
|
67 |
parser.add_argument(
|
68 |
+
"--sample_rate", type=int, default=48000,
|
69 |
+
help="Sample rate for audio output.",
|
70 |
)
|
71 |
parser.add_argument(
|
72 |
"--save_dir", type=str, default="./outputs/tmp",
|
73 |
+
help="output save dir"
|
74 |
)
|
75 |
parser.add_argument(
|
76 |
"--data_path", type=str, default="data/video_processed/video_gt_augment",
|
77 |
+
help="inference data path"
|
78 |
)
|
79 |
|
80 |
args = parser.parse_args()
|
|
|
183 |
mel = vae.decode_first_stage(latents)
|
184 |
wave = vae.decode_to_waveform(mel)
|
185 |
|
186 |
+
all_outputs += [item for item in wave]
|
187 |
|
188 |
# Save #
|
189 |
exp_id = str(int(time.time()))
|
|
|
194 |
output_dir = "{}/{}_{}_steps_{}_guidance_{}_sampleRate_{}_augment".format(args.save_dir, exp_id, "_".join(args.model.split("/")[1:-1]), num_steps, guidance, sample_rate)
|
195 |
os.makedirs(output_dir, exist_ok=True)
|
196 |
for j, wav in enumerate(all_outputs):
|
197 |
+
sf.write("{}/{}".format(output_dir, wavname[j]), wav, samplerate=sample_rate)
|
|
|
|
|
|
|
|
|
|
|
198 |
|
199 |
else:
|
200 |
for i in range(num_samples):
|
|
|
208 |
ranked_wavs_for_text = [wavs_for_text[r] for r in rank]
|
209 |
|
210 |
for i, wav in enumerate(ranked_wavs_for_text):
|
211 |
+
output_dir = "{}/{}_{}_steps_{}_guidance_{}_sampleRate_{}/rank_{}".format(args.save_dir, exp_id, "_".join(args.model.split("/")[1:-1]), num_steps, guidance, sample_rate, i+1)
|
212 |
+
sf.write("{}/{}".format(output_dir, wavname[k]), wav, samplerate=sample_rate)
|
|
|
|
|
|
|
|
|
213 |
|
214 |
if __name__ == "__main__":
|
215 |
+
main()
|