ajayarora1235
commited on
Commit
·
35a24d7
1
Parent(s):
aa9ee54
fix extra dependencies in app
Browse files
app.py
CHANGED
@@ -31,10 +31,6 @@ os.environ["TEMP"] = tmp
|
|
31 |
warnings.filterwarnings("ignore")
|
32 |
torch.manual_seed(114514)
|
33 |
|
34 |
-
from ilariatts import tts_order_voice
|
35 |
-
language_dict = tts_order_voice
|
36 |
-
ilariavoices = language_dict.keys()
|
37 |
-
|
38 |
import signal
|
39 |
|
40 |
import math
|
@@ -1407,9 +1403,6 @@ def mouth(size, face, voice, faces):
|
|
1407 |
process = subprocess.Popen(command, shell=True, cwd='/content/wav2lip-HD/Wav2Lip-master')
|
1408 |
stdout, stderr = process.communicate()
|
1409 |
return '/content/wav2lip-HD/outputs/result.mp4', 'Animation completed.'
|
1410 |
-
eleven_voices = ['Adam','Antoni','Josh','Arnold','Sam','Bella','Rachel','Domi','Elli']
|
1411 |
-
eleven_voices_ids=['pNInz6obpgDQGcFmaJgB','ErXwobaYiN019PkySvjV','TxGEqnHWrfWFTfGW9XjX','VR6AewLTigWG4xSOukaG','yoZ06aMxZJJ28mfd3POQ','EXAVITQu4vr4xnSDxMaL','21m00Tcm4TlvDq8ikWAM','AZnzlk1XvdvUeBnXmlld','MF3mGyEYCl7XYWbV9V6O']
|
1412 |
-
chosen_voice = dict(zip(eleven_voices, eleven_voices_ids))
|
1413 |
|
1414 |
def stoptraining(mim):
|
1415 |
if int(mim) == 1:
|
@@ -1424,53 +1417,7 @@ def stoptraining(mim):
|
|
1424 |
)
|
1425 |
|
1426 |
|
1427 |
-
|
1428 |
-
# if xiapi!= '' and id !='':
|
1429 |
-
# choice = chosen_voice[id]
|
1430 |
-
# CHUNK_SIZE = 1024
|
1431 |
-
# url = f"https://api.elevenlabs.io/v1/text-to-speech/{choice}"
|
1432 |
-
# headers = {
|
1433 |
-
# "Accept": "audio/mpeg",
|
1434 |
-
# "Content-Type": "application/json",
|
1435 |
-
# "xi-api-key": xiapi
|
1436 |
-
# }
|
1437 |
-
# if lang == 'en':
|
1438 |
-
# data = {
|
1439 |
-
# "text": text,
|
1440 |
-
# "model_id": "eleven_monolingual_v1",
|
1441 |
-
# "voice_settings": {
|
1442 |
-
# "stability": 0.5,
|
1443 |
-
# "similarity_boost": 0.5
|
1444 |
-
# }
|
1445 |
-
# }
|
1446 |
-
# else:
|
1447 |
-
# data = {
|
1448 |
-
# "text": text,
|
1449 |
-
# "model_id": "eleven_multilingual_v1",
|
1450 |
-
# "voice_settings": {
|
1451 |
-
# "stability": 0.5,
|
1452 |
-
# "similarity_boost": 0.5
|
1453 |
-
# }
|
1454 |
-
# }
|
1455 |
-
|
1456 |
-
# response = requests.post(url, json=data, headers=headers)
|
1457 |
-
# with open('./temp_eleven.mp3', 'wb') as f:
|
1458 |
-
# for chunk in response.iter_content(chunk_size=CHUNK_SIZE):
|
1459 |
-
# if chunk:
|
1460 |
-
# f.write(chunk)
|
1461 |
-
# aud_path = save_to_wav('./temp_eleven.mp3')
|
1462 |
-
# return aud_path, aud_path
|
1463 |
-
# else:
|
1464 |
-
# tts = gTTS(text, lang=lang)
|
1465 |
-
# tts.save('./temp_gTTS.mp3')
|
1466 |
-
# aud_path = save_to_wav('./temp_gTTS.mp3')
|
1467 |
-
# return aud_path, aud_path
|
1468 |
-
|
1469 |
-
# def ilariaTTS(text, ttsvoice):
|
1470 |
-
# vo=language_dict[ttsvoice]
|
1471 |
-
# asyncio.run(edge_tts.Communicate(text, vo).save("./temp_ilaria.mp3"))
|
1472 |
-
# aud_path = save_to_wav('./temp_ilaria.mp3')
|
1473 |
-
# return aud_path, aud_path
|
1474 |
|
1475 |
def transcribe_btn_click(audio_choice):
|
1476 |
batch_size = 1 # Adjust based on your GPU memory availability
|
@@ -1619,7 +1566,7 @@ for root, dirs, files in os.walk(index_root, topdown=False):
|
|
1619 |
if name.endswith(".index") and "trained" not in name:
|
1620 |
index_paths.append("%s/%s" % (root, name))
|
1621 |
|
1622 |
-
with gr.Blocks(theme=gr.themes.Default(primary_hue="pink", secondary_hue="rose"), title="
|
1623 |
with gr.Tabs():
|
1624 |
with gr.TabItem("Inference"):
|
1625 |
# gr.HTML("<h1> Ilaria RVC 💖 </h1>")
|
@@ -1744,13 +1691,6 @@ with gr.Blocks(theme=gr.themes.Default(primary_hue="pink", secondary_hue="rose")
|
|
1744 |
interactive=False,
|
1745 |
)
|
1746 |
|
1747 |
-
# with gr.Accordion('IlariaTTS', open=True):
|
1748 |
-
# with gr.Column():
|
1749 |
-
# ilariaid=gr.Dropdown(label="Voice:", choices=ilariavoices, value="English-Jenny (Female)")
|
1750 |
-
# ilariatext = gr.Textbox(label="Input your Text", interactive=True, value="This is a test.")
|
1751 |
-
# ilariatts_button = gr.Button(value="Speak")
|
1752 |
-
# ilariatts_button.click(fn=ilariaTTS, inputs=[ilariatext, ilariaid], outputs=[record_button, input_audio0])
|
1753 |
-
|
1754 |
#with gr.Column():
|
1755 |
with gr.Accordion("Index Settings", open=False):
|
1756 |
#with gr.Row():
|
|
|
31 |
warnings.filterwarnings("ignore")
|
32 |
torch.manual_seed(114514)
|
33 |
|
|
|
|
|
|
|
|
|
34 |
import signal
|
35 |
|
36 |
import math
|
|
|
1403 |
process = subprocess.Popen(command, shell=True, cwd='/content/wav2lip-HD/Wav2Lip-master')
|
1404 |
stdout, stderr = process.communicate()
|
1405 |
return '/content/wav2lip-HD/outputs/result.mp4', 'Animation completed.'
|
|
|
|
|
|
|
1406 |
|
1407 |
def stoptraining(mim):
|
1408 |
if int(mim) == 1:
|
|
|
1417 |
)
|
1418 |
|
1419 |
|
1420 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1421 |
|
1422 |
def transcribe_btn_click(audio_choice):
|
1423 |
batch_size = 1 # Adjust based on your GPU memory availability
|
|
|
1566 |
if name.endswith(".index") and "trained" not in name:
|
1567 |
index_paths.append("%s/%s" % (root, name))
|
1568 |
|
1569 |
+
with gr.Blocks(theme=gr.themes.Default(primary_hue="pink", secondary_hue="rose"), title="HITGEN AI") as app:
|
1570 |
with gr.Tabs():
|
1571 |
with gr.TabItem("Inference"):
|
1572 |
# gr.HTML("<h1> Ilaria RVC 💖 </h1>")
|
|
|
1691 |
interactive=False,
|
1692 |
)
|
1693 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1694 |
#with gr.Column():
|
1695 |
with gr.Accordion("Index Settings", open=False):
|
1696 |
#with gr.Row():
|