Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -1,4 +1,4 @@
|
|
1 |
-
|
2 |
import spaces
|
3 |
import soundfile as sf
|
4 |
import torch
|
@@ -12,8 +12,6 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, VitsModel
|
|
12 |
import torch
|
13 |
import numpy as np
|
14 |
import os
|
15 |
-
import argparse
|
16 |
-
import gradio as gr
|
17 |
from timeit import default_timer as timer
|
18 |
import torch
|
19 |
import numpy as np
|
@@ -26,7 +24,6 @@ import whisper
|
|
26 |
# tts_model.to("cuda")
|
27 |
# print("TTS Loaded!")
|
28 |
|
29 |
-
|
30 |
def load_whisper():
|
31 |
return whisper.load_model("medium", device = 'cpu')
|
32 |
|
@@ -100,6 +97,8 @@ def _parse_text(text):
|
|
100 |
lines[i] = "<br>" + line
|
101 |
text = "".join(lines)
|
102 |
return text
|
|
|
|
|
103 |
@spaces.GPU
|
104 |
def predict(_query, _chatbot, _task_history):
|
105 |
print(f"User: {_parse_text(_query)}")
|
@@ -116,6 +115,7 @@ def predict(_query, _chatbot, _task_history):
|
|
116 |
_task_history.append((_query, full_response))
|
117 |
print(f"Qwen-7B-Chat: {_parse_text(full_response)}")
|
118 |
|
|
|
119 |
def read_text(text):
|
120 |
print("___Tekst do przeczytania!")
|
121 |
inputs = tokenizer_tss(text, return_tensors="pt").to("cuda")
|
@@ -127,7 +127,7 @@ def read_text(text):
|
|
127 |
|
128 |
def update_audio(text):
|
129 |
return 'temp_file.wav'
|
130 |
-
|
131 |
def translate(audio):
|
132 |
print("__Wysyłam nagranie do whisper!")
|
133 |
transcription = whisper_model.transcribe(audio, language="pl")
|
|
|
1 |
+
import gradio as gr
|
2 |
import spaces
|
3 |
import soundfile as sf
|
4 |
import torch
|
|
|
12 |
import torch
|
13 |
import numpy as np
|
14 |
import os
|
|
|
|
|
15 |
from timeit import default_timer as timer
|
16 |
import torch
|
17 |
import numpy as np
|
|
|
24 |
# tts_model.to("cuda")
|
25 |
# print("TTS Loaded!")
|
26 |
|
|
|
27 |
def load_whisper():
|
28 |
return whisper.load_model("medium", device = 'cpu')
|
29 |
|
|
|
97 |
lines[i] = "<br>" + line
|
98 |
text = "".join(lines)
|
99 |
return text
|
100 |
+
|
101 |
+
|
102 |
@spaces.GPU
|
103 |
def predict(_query, _chatbot, _task_history):
|
104 |
print(f"User: {_parse_text(_query)}")
|
|
|
115 |
_task_history.append((_query, full_response))
|
116 |
print(f"Qwen-7B-Chat: {_parse_text(full_response)}")
|
117 |
|
118 |
+
@spaces.GPU
|
119 |
def read_text(text):
|
120 |
print("___Tekst do przeczytania!")
|
121 |
inputs = tokenizer_tss(text, return_tensors="pt").to("cuda")
|
|
|
127 |
|
128 |
def update_audio(text):
|
129 |
return 'temp_file.wav'
|
130 |
+
@spaces.GPU
|
131 |
def translate(audio):
|
132 |
print("__Wysyłam nagranie do whisper!")
|
133 |
transcription = whisper_model.transcribe(audio, language="pl")
|