Update app.py

app.py

@@ -1,92 +1 @@
-# -*- coding: utf-8 -*-
-"""gradio_kurdi.ipynb
-
-Automatically generated by Colab.
-
-Original file is located at
-    https://colab.research.google.com/drive/1DFSu80KU5dHKmbbqb2bPA5R8hzPbTP76
-"""
-
-!pip install torch transformers datasets librosa gradio
-
-import torch
-from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
-import librosa
-import numpy as np
-from datetime import timedelta
-import gradio as gr
-import os
-
-def format_time(seconds):
-    td = timedelta(seconds=seconds)
-    hours, remainder = divmod(td.seconds, 3600)
-    minutes, seconds = divmod(remainder, 60)
-    milliseconds = td.microseconds // 1000
-    return f"{hours:02d}:{minutes:02d}:{seconds:02d},{milliseconds:03d}"
-
-def estimate_word_timings(transcription, total_duration):
-    words = transcription.split()
-    total_chars = sum(len(word) for word in words)
-    char_duration = total_duration / total_chars
-
-    word_timings = []
-    current_time = 0
-
-    for word in words:
-        word_duration = len(word) * char_duration
-        start_time = current_time
-        end_time = current_time + word_duration
-        word_timings.append((word, start_time, end_time))
-        current_time = end_time
-
-    return word_timings
-
-model_name = "Akashpb13/xlsr_kurmanji_kurdish"
-model = Wav2Vec2ForCTC.from_pretrained(model_name)
-processor = Wav2Vec2Processor.from_pretrained(model_name)
-
-def transcribe_audio(file):
-    # معالجة الملف الصوتي
-    speech, rate = librosa.load(file, sr=16000)
-
-    # تحضير البيانات
-    input_values = processor(speech, return_tensors="pt", sampling_rate=rate).input_values
-
-    # تنبؤات النموذج
-    with torch.no_grad():
-        logits = model(input_values).logits
-
-    # الحصول على النص من التنبؤات
-    predicted_ids = torch.argmax(logits, dim=-1)
-    transcription = processor.batch_decode(predicted_ids)[0]
-
-    # حساب الوقت الإجمالي للتسجيل
-    total_duration = len(speech) / rate
-
-    # تقدير توقيت كل كلمة
-    word_timings = estimate_word_timings(transcription, total_duration)
-
- # إنشاء محتوى ملف SRT
-    srt_content = ""
-    for i, (word, start_time, end_time) in enumerate(word_timings, start=1):
-        start_time_str = format_time(start_time)
-        end_time_str = format_time(end_time)
-        srt_content += f"{i}\n{start_time_str} --> {end_time_str}\n{word}\n\n"
-
-    # حفظ الملف SRT
-    output_filename = "output_word_by_word.srt"
-    with open(output_filename, "w", encoding="utf-8") as f:
-        f.write(srt_content)
-
-    return transcription, output_filename
-
-interface = gr.Interface(
-    fn=transcribe_audio,
-    inputs=gr.Audio(type="filepath"),
-    outputs=[gr.Textbox(label="Transcription"), gr.File(label="Download SRT File")],
-    title="Deng --- Nivîsandin ::: Kurdî-Kurmancî",
-    description="Dengê xwe ji me re rêke û li Submit bixe ... û bila bêhna te fireh be .",
-     article="By Derax Elî"
-)
-
-interface.launch()
+https://89879aa8bd9cacb387.gradio.live