Spaces:
Running
Running
Fix LM assert error
Browse files- app.py +2 -2
- audiocraft/models/lm.py +2 -1
app.py
CHANGED
@@ -266,14 +266,14 @@ def predict(model, text, melody_filepath, duration, dimension, topk, topp, tempe
|
|
266 |
descriptions=[text],
|
267 |
melody_wavs=melody,
|
268 |
melody_sample_rate=sr,
|
269 |
-
progress=
|
270 |
)
|
271 |
# All output_segments are populated, so we can break the loop or set duration to 0
|
272 |
break
|
273 |
else:
|
274 |
#output = MODEL.generate(descriptions=[text], progress=False)
|
275 |
if not output_segments:
|
276 |
-
next_segment = MODEL.generate(descriptions=[text], progress=
|
277 |
duration -= segment_duration
|
278 |
else:
|
279 |
last_chunk = output_segments[-1][:, :, -overlap*MODEL.sample_rate:]
|
|
|
266 |
descriptions=[text],
|
267 |
melody_wavs=melody,
|
268 |
melody_sample_rate=sr,
|
269 |
+
progress=False
|
270 |
)
|
271 |
# All output_segments are populated, so we can break the loop or set duration to 0
|
272 |
break
|
273 |
else:
|
274 |
#output = MODEL.generate(descriptions=[text], progress=False)
|
275 |
if not output_segments:
|
276 |
+
next_segment = MODEL.generate(descriptions=[text], progress=False)
|
277 |
duration -= segment_duration
|
278 |
else:
|
279 |
last_chunk = output_segments[-1][:, :, -overlap*MODEL.sample_rate:]
|
audiocraft/models/lm.py
CHANGED
@@ -456,7 +456,8 @@ class LMModel(StreamingModule):
|
|
456 |
|
457 |
B, K, T = prompt.shape
|
458 |
start_offset = T
|
459 |
-
|
|
|
460 |
|
461 |
pattern = self.pattern_provider.get_pattern(max_gen_len)
|
462 |
# this token is used as default value for codes that are not generated yet
|
|
|
456 |
|
457 |
B, K, T = prompt.shape
|
458 |
start_offset = T
|
459 |
+
print(f"start_offset: {start_offset} | max_gen_len: {max_gen_len}")
|
460 |
+
assert start_offset <= max_gen_len
|
461 |
|
462 |
pattern = self.pattern_provider.get_pattern(max_gen_len)
|
463 |
# this token is used as default value for codes that are not generated yet
|