Spaces:
Runtime error
Runtime error
XiaoHei Studio
commited on
Commit
•
41f245a
1
Parent(s):
2d30eab
Update app.py
Browse files
app.py
CHANGED
@@ -434,9 +434,9 @@ def vc_infer(output_format, sid, input_audio, sr, input_audio_path, vc_transform
|
|
434 |
input_audio = librosa.to_mono(input_audio.transpose(1, 0))
|
435 |
if sr != 44100:
|
436 |
input_audio = librosa.resample(input_audio, orig_sr=sr, target_sr=44100)
|
437 |
-
sf.write("
|
438 |
_audio = model.slice_inference(
|
439 |
-
"
|
440 |
sid,
|
441 |
vc_transform,
|
442 |
slice_db,
|
@@ -490,7 +490,7 @@ def vc_fn(output_format, sid, input_audio, vc_transform, auto_f0, cluster_ratio,
|
|
490 |
return "你还未加载聚类或特征检索模型,无法启用聚类/特征检索混合比例", None
|
491 |
audio, sr = sf.read(input_audio)
|
492 |
output_file_path = vc_infer(output_format, sid, audio, sr, input_audio, vc_transform, auto_f0, cluster_ratio, slice_db, noise_scale, pad_seconds, cl_num, lg_num, lgr_num, f0_predictor, enhancer_adaptive_key, cr_threshold, k_step, use_spk_mix, second_encoding, loudness_envelope_adjustment)
|
493 |
-
os.remove("
|
494 |
return "Success", output_file_path
|
495 |
except Exception as e:
|
496 |
if debug:
|
@@ -536,13 +536,13 @@ def tts_fn(_text, _gender, _lang, _rate, _volume, output_format, sid, vc_transfo
|
|
536 |
else:
|
537 |
subprocess.run([r"python", "tts.py", _text, _lang, _rate, _volume])
|
538 |
target_sr = 44100
|
539 |
-
y, sr = librosa.load("
|
540 |
resampled_y = librosa.resample(y, orig_sr=sr, target_sr=target_sr)
|
541 |
-
sf.write("
|
542 |
-
input_audio = "
|
543 |
audio, sr = sf.read(input_audio)
|
544 |
output_file_path = vc_infer(output_format, sid, audio, sr, input_audio, vc_transform, auto_f0, cluster_ratio, slice_db, noise_scale, pad_seconds, cl_num, lg_num, lgr_num, f0_predictor, enhancer_adaptive_key, cr_threshold, k_step, use_spk_mix, second_encoding, loudness_envelope_adjustment)
|
545 |
-
#os.remove("
|
546 |
return "Success", output_file_path
|
547 |
except Exception as e:
|
548 |
if debug:
|
|
|
434 |
input_audio = librosa.to_mono(input_audio.transpose(1, 0))
|
435 |
if sr != 44100:
|
436 |
input_audio = librosa.resample(input_audio, orig_sr=sr, target_sr=44100)
|
437 |
+
sf.write("temp.wav", input_audio, 44100, format="wav")
|
438 |
_audio = model.slice_inference(
|
439 |
+
"temp.wav",
|
440 |
sid,
|
441 |
vc_transform,
|
442 |
slice_db,
|
|
|
490 |
return "你还未加载聚类或特征检索模型,无法启用聚类/特征检索混合比例", None
|
491 |
audio, sr = sf.read(input_audio)
|
492 |
output_file_path = vc_infer(output_format, sid, audio, sr, input_audio, vc_transform, auto_f0, cluster_ratio, slice_db, noise_scale, pad_seconds, cl_num, lg_num, lgr_num, f0_predictor, enhancer_adaptive_key, cr_threshold, k_step, use_spk_mix, second_encoding, loudness_envelope_adjustment)
|
493 |
+
os.remove("temp.wav")
|
494 |
return "Success", output_file_path
|
495 |
except Exception as e:
|
496 |
if debug:
|
|
|
536 |
else:
|
537 |
subprocess.run([r"python", "tts.py", _text, _lang, _rate, _volume])
|
538 |
target_sr = 44100
|
539 |
+
y, sr = librosa.load("tts.wav")
|
540 |
resampled_y = librosa.resample(y, orig_sr=sr, target_sr=target_sr)
|
541 |
+
sf.write("tts.wav", resampled_y, target_sr, subtype = "PCM_16")
|
542 |
+
input_audio = "tts.wav"
|
543 |
audio, sr = sf.read(input_audio)
|
544 |
output_file_path = vc_infer(output_format, sid, audio, sr, input_audio, vc_transform, auto_f0, cluster_ratio, slice_db, noise_scale, pad_seconds, cl_num, lg_num, lgr_num, f0_predictor, enhancer_adaptive_key, cr_threshold, k_step, use_spk_mix, second_encoding, loudness_envelope_adjustment)
|
545 |
+
#os.remove("tts.wav")
|
546 |
return "Success", output_file_path
|
547 |
except Exception as e:
|
548 |
if debug:
|