import gradio as gr
import requests
import json
import os

BASE_URL = "https://api.jigsawstack.com/v1"
headers = {
    "x-api-key": os.getenv("JIGSAWSTACK_API_KEY")
}

def transcribe_audio(input_type, audio_url, file_store_key, language):
    """Transcribe audio using JigsawStack Speech-to-Text API"""
    if input_type == "Audio URL" and not audio_url:
        return "Error: Please provide an audio URL.", ""
    if input_type == "File Store Key" and not file_store_key:
        return "Error: Please provide a file store key.", ""
    try:
        payload = {}
        if input_type == "Audio URL":
            payload["url"] = audio_url.strip()
        if input_type == "File Store Key":
            payload["file_store_key"] = file_store_key.strip()
        if language:
            payload["language"] = language
        response = requests.post(
            f"{BASE_URL}/ai/transcribe",
            headers=headers,
            json=payload
        )
        response.raise_for_status()
        result = response.json()
        if not result.get("success"):
            error_msg = f"Error: API call failed - {result.get('message', 'Unknown error')}"
            return error_msg, ""
        transcribed_text = result.get("text", "")
        return "Transcription completed successfully!", transcribed_text
    except requests.exceptions.RequestException as e:
        return f"Request failed: {str(e)}", ""
    except Exception as e:
        return f"An unexpected error occurred: {str(e)}", ""

with gr.Blocks() as demo:
    gr.Markdown("""
        <div style='text-align: center; margin-bottom: 24px;'>
            <h1 style='font-size:2.2em; margin-bottom: 0.2em;'>Speech-to-Text Transcription</h1>
            <p style='font-size:1.2em; margin-top: 0;'>Transcribe video and audio files with ease leveraging Whisper large V3 AI model.</p>
            <p style='font-size:1em; margin-top: 0.5em;'>Supported formats: MP3, WAV, M4A, FLAC, AAC, OGG, WEBM. Max file size: 100MB, Max duration: 4 hours.</p>
        </div>
        """)
    with gr.Row():
        with gr.Column():
            gr.Markdown("#### Audio Input")
            input_type = gr.Radio([
                "Audio URL",
                "File Store Key"
            ], value="Audio URL", label="Select Input Type")
            audio_url = gr.Textbox(
                label="Audio URL",
                placeholder="Enter the URL of the audio/video file...",
                visible=True
            )
            file_store_key = gr.Textbox(
                label="File Store Key",
                placeholder="Enter the file store key from JigsawStack File Storage...",
                visible=False
            )
            language = gr.Textbox(
                label="Language (optional)",
                placeholder="e.g., en, es, fr, de, ja, zh... (leave empty for auto-detect)"
            )
            transcribe_btn = gr.Button("Start Transcription", variant="primary")
        with gr.Column():
            gr.Markdown("#### Transcription Result")
            status_message = gr.Textbox(label="Status", interactive=False)
            transcribed_text = gr.Textbox(
                label="Transcribed Text",
                interactive=False,
                lines=10,
                max_lines=20
            )
    def toggle_inputs(selected):
        if selected == "Audio URL":
            return gr.update(visible=True), gr.update(visible=False)
        else:
            return gr.update(visible=False), gr.update(visible=True)
    input_type.change(toggle_inputs, inputs=[input_type], outputs=[audio_url, file_store_key])
    transcribe_btn.click(
        transcribe_audio,
        inputs=[input_type, audio_url, file_store_key, language],
        outputs=[status_message, transcribed_text]
    )

demo.launch()