Spaces:
Runtime error
Runtime error
File size: 2,078 Bytes
474f4cc cc84c76 1197cf0 9f8e59c 6d916f4 474f4cc 9e1c381 1bfa7fa 4b5b0e4 ac5edc7 9f8e59c 29a0691 ad135b5 29a0691 de0a7fd 29a0691 589c7da 6d916f4 0b298e3 ac5edc7 cc84c76 0b298e3 e2bb816 b56acb6 0b298e3 29382d3 0acfb8a 29382d3 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 |
import streamlit as st
from gtts import gTTS
from io import BytesIO
import torch
import torchaudio
from PyPDF2 import PdfReader
from whisperspeech.pipeline import Pipeline
pipe = Pipeline(torch_compile=True)
st.image('OIG3 (4).jpeg', caption='Your host on this PDF-to-Speech adventure!')
x = st.slider('Select the number of pages you wish to transcribe')
uploaded_file = st.file_uploader("Choose a file", "pdf")
if uploaded_file is not None:
# creating a pdf reader object
reader = PdfReader(uploaded_file)
# printing number of pages in pdf file
X = len(reader.pages)
print(X)
i = 0
while i <= X and i <= x:
# getting a specific page from the pdf file
page = reader.pages[i]
# extracting text from page
text = page.extract_text()
print("Created text of page", i )
# Generate audio for the current page using a unique filename
page_audio_file = f"output_{i}.wav"
pipe.generate_to_file(page_audio_file, text)
# Display the generated audio using st.audio
with open(page_audio_file, "rb") as audio_file:
st.audio(audio_file)
# sound_file = BytesIO()
# tts = gTTS(text, lang='en')
# tts.write_to_fp(sound_file)
# st.audio(sound_file)
print("Read aloud", i, "pages of", X, "total pages.")
i = i + 1
st.write("🎉 That's the whole PDF! Have an awesome day! 🎉")
prompt = st.chat_input("Copy/Paste or type in text to have read aloud")
if prompt:
st.write(prompt)
with st.popover("✨ Open your text-to-speech from text input ✨"):
#sound_file = BytesIO()
#tts = gTTS(prompt, lang='en')
#tts.write_to_fp(sound_file)
#st.audio(sound_file)
# Generate audio for the current page using a unique filename
page_audio_file = f"output_{i}.wav"
pipe.generate_to_file(page_audio_file, text)
# Display the generated audio using st.audio
with open(page_audio_file, "rb") as audio_file:
st.audio(audio_file)
|