rdlf commited on
Commit
46a6ba3
·
verified ·
1 Parent(s): 2aad103

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +5 -9
app.py CHANGED
@@ -1,13 +1,10 @@
1
  import streamlit as st
2
  from PIL import Image
3
  import os
4
- from openai import OpenAI, Whisper
5
 
6
  st.image('calamo.png', caption="", use_column_width=False)
7
 
8
- # Initialize OpenAI and Whisper
9
- openai = OpenAI(api_key=os.environ.get("OPENAI_API_KEY"))
10
- whisper = Whisper(openai)
11
 
12
  client = Groq(
13
  api_key=os.environ.get("GROQ_API_KEY"),
@@ -42,10 +39,9 @@ def process_text(input_text):
42
 
43
  def generate_audio(input_text):
44
  tts = process_text(input_text)
45
- speech = whisper.synthesize(tts)
46
- with open("whisper_out.wav", "wb") as f:
47
- f.write(speech.audio)
48
- return "whisper_out.wav"
49
 
50
  user_input = st.text_input("Pega un texto para aclararlo y escuchar una lectura.")
51
 
@@ -54,4 +50,4 @@ if st.button('Aclarar'):
54
  st.write("Pega un texto aquí")
55
  else:
56
  speech_file = generate_audio(user_input)
57
- st.audio(speech_file, format='audio/wav')
 
1
  import streamlit as st
2
  from PIL import Image
3
  import os
4
+ from gtts import gTTS
5
 
6
  st.image('calamo.png', caption="", use_column_width=False)
7
 
 
 
 
8
 
9
  client = Groq(
10
  api_key=os.environ.get("GROQ_API_KEY"),
 
39
 
40
  def generate_audio(input_text):
41
  tts = process_text(input_text)
42
+ speech = gTTS(text=tts, lang='es', slow=False)
43
+ speech.save("gtts_out.mp3")
44
+ return "gtts_out.mp3"
 
45
 
46
  user_input = st.text_input("Pega un texto para aclararlo y escuchar una lectura.")
47
 
 
50
  st.write("Pega un texto aquí")
51
  else:
52
  speech_file = generate_audio(user_input)
53
+ st.audio(speech_file, format='audio/mp3')