Spaces:
Running
Running
Update app_rvc.py
Browse files- app_rvc.py +34 -3
app_rvc.py
CHANGED
@@ -4,6 +4,7 @@ os.system("pip install -q piper-tts==1.2.0")
|
|
4 |
os.system("pip install -q -r requirements_xtts.txt")
|
5 |
os.system("pip install -q TTS==0.21.1 --no-deps")
|
6 |
import spaces
|
|
|
7 |
from soni_translate.logging_setup import (
|
8 |
logger,
|
9 |
set_logging_level,
|
@@ -661,6 +662,31 @@ class SoniTranslate(SoniTrCache):
|
|
661 |
logger.info(f"Done: {output}")
|
662 |
return output
|
663 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
664 |
if not self.task_in_cache("refine_vocals", [vocal_refinement], {}):
|
665 |
self.vocals = None
|
666 |
if vocal_refinement:
|
@@ -1308,9 +1334,14 @@ class SoniTranslate(SoniTrCache):
|
|
1308 |
if not document:
|
1309 |
raise Exception("No data found")
|
1310 |
|
1311 |
-
if os.environ.get("
|
1312 |
-
raise RuntimeError(
|
1313 |
-
|
|
|
|
|
|
|
|
|
|
|
1314 |
if "videobook" in output_type:
|
1315 |
if not document.lower().endswith(".pdf"):
|
1316 |
raise ValueError(
|
|
|
4 |
os.system("pip install -q -r requirements_xtts.txt")
|
5 |
os.system("pip install -q TTS==0.21.1 --no-deps")
|
6 |
import spaces
|
7 |
+
import librosa
|
8 |
from soni_translate.logging_setup import (
|
9 |
logger,
|
10 |
set_logging_level,
|
|
|
662 |
logger.info(f"Done: {output}")
|
663 |
return output
|
664 |
|
665 |
+
if os.environ.get("IS_DEMO") == "TRUE":
|
666 |
+
duration_verify = librosa.get_duration(filename=base_audio_wav)
|
667 |
+
logger.info(f"Duration: {duration_verify} seconds")
|
668 |
+
if duration_verify > 1500:
|
669 |
+
raise RuntimeError(
|
670 |
+
"The audio is too long to process in this demo. Alternatively, you"
|
671 |
+
" can install the app locally or use the Colab notebook available "
|
672 |
+
"in the SoniTranslate repository."
|
673 |
+
)
|
674 |
+
elif duration_verify > 300:
|
675 |
+
tts_voices_list = [
|
676 |
+
tts_voice00, tts_voice01, tts_voice02, tts_voice03, tts_voice04,
|
677 |
+
tts_voice05, tts_voice06, tts_voice07, tts_voice08, tts_voice09,
|
678 |
+
tts_voice10, tts_voice11
|
679 |
+
]
|
680 |
+
|
681 |
+
for tts_voice_ in tts_voices_list:
|
682 |
+
if "_XTTS_" in tts_voice_:
|
683 |
+
raise RuntimeError(
|
684 |
+
"XTTS is too slow to be used for audio longer than 5 "
|
685 |
+
"minutes in this demo. Alternatively, you can install "
|
686 |
+
"the app locally or use the Colab notebook available in"
|
687 |
+
" the SoniTranslate repository."
|
688 |
+
)
|
689 |
+
|
690 |
if not self.task_in_cache("refine_vocals", [vocal_refinement], {}):
|
691 |
self.vocals = None
|
692 |
if vocal_refinement:
|
|
|
1334 |
if not document:
|
1335 |
raise Exception("No data found")
|
1336 |
|
1337 |
+
if os.environ.get("IS_DEMO") == "TRUE" and not is_string:
|
1338 |
+
raise RuntimeError(
|
1339 |
+
"This option is disabled in this demo. "
|
1340 |
+
"Alternatively, you can install "
|
1341 |
+
"the app locally or use the Colab notebook available in"
|
1342 |
+
" the SoniTranslate repository."
|
1343 |
+
)
|
1344 |
+
|
1345 |
if "videobook" in output_type:
|
1346 |
if not document.lower().endswith(".pdf"):
|
1347 |
raise ValueError(
|