import gradio as gr
from transformers import pipeline
classifier = pipeline('sentiment-analysis')
#################################################################
#1: Image to text
from OCR_Image_to_Text import get_OCR_demo

#################################################################
#2: Text to Speech
title = "Text to Speech Translation"
tts_examples = [
    "I love learning machine learning",
    "How do you do?",
]
tts_demo = gr.Interface.load(
    "huggingface/facebook/fastspeech2-en-ljspeech",
    title = title,
    examples=tts_examples,
    description="Give me something to say!",
)

#################################################################
# #2_1: Text to Speech_1
title = "Text FILE to Speech Translation"
import gradio as gr
from gtts import gTTS
from io import BytesIO

def text_to_speech(file):
    # Create a gTTS (Google Text-to-Speech) object
    with open(file.name, 'r') as f:
        text = f.read()
    tts = gTTS(text)
    # Save the speech as an audio file in memory
    audio_stream = BytesIO()
    tts.save(audio_stream)
    audio_stream.seek(0)
    return audio_stream
    file.close()
  
inputs = gr.inputs.File(label="Upload a Text File")
outputs = gr.outputs.Audio(label="Audio")

interface_demo = gr.Interface.load("huggingface/facebook/fastspeech2-en-ljspeech",fn=text_to_speech,title = title, inputs=inputs, outputs=outputs,description="Give me something to say!",capture_session=True)


#################################################################
#3: Sentiment Analysis
def get_sentiment(input_text):
  return classifier(input_text)
  
sentiment_demo= gr.Interface(fn=get_sentiment,  inputs = 'text',outputs=['text'],title = 'Sentiment Analysis', description = "Enter a sentence and know about it's sentiment",examples = [
    ["We are very happy to show you the 🤗 Transformers library."],
    ["I am happy with the performance of Indian Hockey team"],
    ["Pizza was not at all hot and restaurant was having the pathetic service"]
])
#sentiment_demo.launch(inline = False)
#################################################################    
#4 POS Tagging
import nltk
nltk.download('punkt')
nltk.download('averaged_perceptron_tagger')
def get_POS(input_text):
  sent_tag = nltk.word_tokenize(input_text)
  for token in sent_tag:

    return nltk.pos_tag([token])

title = 'Know about Parts of Speech (POS) Tags'
examples = [
    ["We are very happy to show you the 🤗 Transformers library."],
    ["I am happy with the performance of Indian Hockey team"],
    ["Pizza was not at all hot and restaurant was having the pathetic service"],
    ["Scientist Dr. Evangeline Starlight of Technopolish announced a breakthrough in quantum computing at Nova University. Mayor Orion Pulsor commended her. The discovery will be shared at the Galactic Quantum Computing Symposium in Cosmos"]
]
POS_demo = gr.Interface(fn=get_POS, inputs = 'text',outputs=['text'],title = title,description = 'Get POS tags',examples = examples)
#POS_demo.launch(debug=True)

#################################################################
#5 Language Translation
from language_translation import translation_demo


#################################################################
#6: Gradio ASR
from SpeechTranscription_ASR import ST_ASR_demo

#################################################################
#7 YouTube to Text Script generation

from Text_Script_from_YouTube import textScript_demo

########################################################################################################
demo = gr.TabbedInterface([get_OCR_demo, tts_demo, interface_demo, sentiment_demo, POS_demo, translation_demo, ST_ASR_demo, textScript_demo], 
                          ["Image to Text","Text to Speech","Text file to Speech", "Sentiment Analysis","POS findings", "Language Translation","ASR", "YouTube video to Text Script"])
if __name__ == "__main__":
    demo.launch()