import streamlit as st
import scipy

from transformers import AutoProcessor, BarkModel
import os
os.environ["SUNO_OFFLOAD_CPU"] = "True"
os.environ["SUNO_USE_SMALL_MODELS"] = "True"

from bark import SAMPLE_RATE, generate_audio, preload_models
from scipy.io.wavfile import write as write_wav

from transformers import AutoProcessor, BarkModel


print("Loading models")
processor = AutoProcessor.from_pretrained("suno/bark")
model = BarkModel.from_pretrained("suno/bark")
# download and load all models
#preload_models()

#from streamlit_autorefresh import st_autorefresh

# Run the autorefresh about every 2000 milliseconds (2 seconds) and stop
# after it's been refreshed 100 times.
#count = st_autorefresh(interval=2000, limit=100, key="fizzbuzzcounter")

st.write('Welcome to Text to Audio generator app')

def generate_audio_from_text(message):
  audio_array = generate_audio(message)
  scipy.io.wavfile.write("audio.wav", rate=SAMPLE_RATE, data=audio_array)

def generate_audio_2(message):


  voice_preset = "v2/en_speaker_6"

  inputs = processor(message, voice_preset=voice_preset)

  audio_array = model.generate(**inputs)
  audio_array = audio_array.numpy().squeeze()
  sample_rate = model.generation_config.sample_rate
  scipy.io.wavfile.write("audio.wav", rate=sample_rate, data=audio_array)

sentence = st.text_input('Input your sentence here:')

if 'clicked' not in st.session_state:
    st.session_state.clicked = False

def click_button():
  st.session_state.clicked = True
  generate_audio_2(sentence)
  audio_file = open('audio.wav','rb') #enter the filename with filepath
  audio_bytes = audio_file.read() #reading the file
  st.audio(audio_bytes, format='audio/wav') #displaying the audio


st.button('Click me', on_click=click_button)