update all

app.py

@@ -1,6 +1,8 @@
 from image import *
 from language import *
+from translation import *
 import streamlit as st
+import os
 
 
 # url = "https://i.imgur.com/qs0CxjE_d.webp?maxwidth=760&fidelity=grand"
@@ -14,8 +16,9 @@ url = st.text_input('mettez le liens de votre image')
 if st.button('générer'):
     responseBase = image(url, question)
     st.write('response is :', responseBase)
+
     st.write('Part 2')
-    print(question)
     st.write(longText(responseBase, question))
 
-print('#### TEST 2####')
+    print("####TEST TTS####")
+    os.system("text-to-speach.py")

language.py

@@ -3,12 +3,11 @@ from transformers import T5Tokenizer, T5ForConditionalGeneration
 def longText(answere, question):
     print('###### LANGUAGES ######')
 
-    input_text = " i have a question and answer.\nthe question is : {}\n the response is : {}\n with this information, can you create an answer phrase?".format(question, answere)
+    input_text = "i have a question and answer.\nthe question is : {}\n the response is : {}\n with this information, can you create an answer phrase?".format(question, answere)
 
     tokenizer = T5Tokenizer.from_pretrained("google/flan-t5-large")
-    model = T5ForConditionalGeneration.from_pretrained("google/flan-t5-large")
-    input_ids = tokenizer(input_text, return_tensors="pt").input_ids
+    model = T5ForConditionalGeneration.from_pretrained("google/flan-t5-large", device_map="auto")
+    input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to("cuda")
     outputs = model.generate(input_ids)
 
-    print(tokenizer.decode(outputs[0]))
     return tokenizer.decode(outputs[0]).replace("<pad>","").replace("</s>","")

requirements.txt

@@ -1,4 +1,5 @@
 transformers[torch]
-torch
-SentencePiece
-sacremoses
+sacremoses
+bark
+scipy
+IPython

text-to-speach.py

@@ -0,0 +1,19 @@
+from bark import SAMPLE_RATE, generate_audio, preload_models
+from scipy.io.wavfile import write as write_wav
+from IPython.display import Audio
+
+# download and load all models
+preload_models()
+
+# generate audio from text
+text_prompt = """
+     Hello, my name is Suno. And, uh — and I like pizza. [laughs] 
+     But I also have other interests such as playing tic tac toe.
+"""
+audio_array = generate_audio(text_prompt)
+
+# save audio to disk
+write_wav("bark_generation.wav", SAMPLE_RATE, audio_array)
+
+# play text in notebook
+Audio(audio_array, rate=SAMPLE_RATE)