import gradio as gr
import azure.cognitiveservices.speech as speechsdk
import time

# Azure credentials
SPEECH_KEY = "vkv2VVTi1agTmU74Sz8C62mOymEHmoknwCaQEnEsroK1AE0B7xt9JQQJ99BDACI8hq2XJ3w3AAAYACOGrzMV"
SERVICE_REGION = "switzerlandnorth"

# Define the language and dialect mapping
language_dialects = {
    "Arabic": {
        "Egypt": "ar-EG",
        "Saudi Arabia": "ar-SA",
        "United Arab Emirates": "ar-AE",
        "Bahrain": "ar-BH",
        "Algeria": "ar-DZ",
        "Iraq": "ar-IQ",
        "Jordan": "ar-JO",
        "Kuwait": "ar-KW",
        "Lebanon": "ar-LB",
        "Libya": "ar-LY",
        "Morocco": "ar-MA",
        "Oman": "ar-OM",
        "Palestinian Authority": "ar-PS",
        "Qatar": "ar-QA",
        "Syria": "ar-SY",
        "Tunisia": "ar-TN",
        "Yemen": "ar-YE"
    },
    "English": {
        "United States": "en-US",
        "United Kingdom": "en-GB",
        "Australia": "en-AU",
        "Canada": "en-CA",
        "India": "en-IN",
        "Ireland": "en-IE",
        "New Zealand": "en-NZ",
        "South Africa": "en-ZA",
        "Singapore": "en-SG",
        "Philippines": "en-PH"
    },
    "French": {
        "France": "fr-FR",
        "Canada": "fr-CA",
        "Switzerland": "fr-CH"
    },
    "Spanish": {
        "Spain": "es-ES",
        "Mexico": "es-MX",
        "Argentina": "es-AR",
        "Colombia": "es-CO",
        "Chile": "es-CL",
        "Peru": "es-PE",
        "Venezuela": "es-VE"
    },
    "German": {
        "Germany": "de-DE",
        "Austria": "de-AT",
        "Switzerland": "de-CH"
    },
    "Portuguese": {
        "Portugal": "pt-PT",
        "Brazil": "pt-BR"
    },
    "Chinese": {
        "Mainland China": "zh-CN",
        "Hong Kong": "zh-HK",
        "Taiwan": "zh-TW"
    },
    "Italian": {
        "Italy": "it-IT"
    },
    "Japanese": {
        "Japan": "ja-JP"
    },
    "Korean": {
        "Korea": "ko-KR"
    }
    # Add more languages and dialects as needed
}


# Function to get dialects based on selected language
def get_dialects(language):
    dialects = list(language_dialects.get(language, {}).keys())
    return gr.update(choices=dialects, value=dialects[0] if dialects else None)

# Function to transcribe audio
def transcribe_audio(audio_file, duration, language, dialect):
    # Simulate recording duration
    print(f"Recording for {duration} seconds...")
    time.sleep(duration)

    # Get the locale code
    locale_code = language_dialects.get(language, {}).get(dialect, "en-US")
    print(f"Selected Locale Code: {locale_code}")

    # Set up speech recognition
    speech_config = speechsdk.SpeechConfig(subscription=SPEECH_KEY, region=SERVICE_REGION)
    speech_config.speech_recognition_language = locale_code
    print(locale_code)
    audio_input = speechsdk.audio.AudioConfig(filename=audio_file)
    speech_recognizer = speechsdk.SpeechRecognizer(speech_config=speech_config, audio_config=audio_input)

    result = speech_recognizer.recognize_once()

    if result.reason == speechsdk.ResultReason.RecognizedSpeech:
        return result.text
    elif result.reason == speechsdk.ResultReason.NoMatch:
        return "No speech could be recognized"
    elif result.reason == speechsdk.ResultReason.Canceled:
        cancellation_details = result.cancellation_details
        return f"Speech recognition canceled: {cancellation_details.error_details}"
    else:
        return "Unknown error occurred during speech recognition"

# Create the Gradio interface
with gr.Blocks() as demo:
    gr.Markdown("## Azure Speech to Text with Language and Dialect Selection")

    with gr.Row():
        audio_input = gr.Audio(type="filepath", label="Upload Audio")
        duration_input = gr.Dropdown(choices=[5, 10], label="Recording Duration", value=5)

    with gr.Row():
        language_input = gr.Dropdown(choices=list(language_dialects.keys()), label="Select Language")
        dialect_input = gr.Dropdown(choices=[], label="Select Dialect")

    transcribe_button = gr.Button("Transcribe")
    output_text = gr.Textbox(label="Transcription Result")

    # Update dialect options based on selected language
    language_input.change(fn=get_dialects, inputs=language_input, outputs=dialect_input)

    # Transcribe audio on button click
    transcribe_button.click(fn=transcribe_audio, inputs=[audio_input, duration_input, language_input, dialect_input], outputs=output_text)

# Launch the app
demo.launch()