Generowanie lektora z napisów

Witam

Mam problem z skryptem do elevenlabs.io

Głos nie generuje dźwięku tak jak wskazuje na plik z napisami srt

Czy ktoś już przerabiał temat ?

import os
import tempfile
import argparse
import requests
from pysubparser import parser
from pydub import AudioSegment
import io
from datetime import datetime, timedelta
import re
import subprocess

CHUNK_SIZE = 1024
ELEVENLAB_API_KEY = 'KLUCZ_API'  # Zastąp swoim kluczem API Elevenlab
VOICE_ID = 'flq6f7yk4E4fJM5XTYuZ'  # Zastąp odpowiednim identyfikatorem głosu

def time_to_seconds(time):
    return (time.hour * 3600 + time.minute * 60 + time.second + time.microsecond / 1000000) if time else 0

def generate_audio(path, rate=200):
    print("Generating audio files for {} with Elevenlab".format(path))

    subtitles = list(parser.parse(path))
    total_subtitles = len(subtitles)

    output_folder = "audio_files"
    os.makedirs(output_folder, exist_ok=True)

    audio_segments = []

    for i, subtitle in enumerate(subtitles):
        text = subtitle.text
        start_time = time_to_seconds(subtitle.start)
        end_time = time_to_seconds(subtitle.end)

        try:
            url = f"https://api.elevenlabs.io/v1/text-to-speech/{VOICE_ID}"

            headers = {
                "Accept": "audio/mpeg",
                "Content-Type": "application/json",
                "xi-api-key": ELEVENLAB_API_KEY
            }

            data = {
                "text": text,
                "model_id": "eleven_multilingual_v2",
                "voice_settings": {
                    "stability": 0.6,
                    "similarity_boost": 0.8
                }
            }

            response = requests.post(url, json=data, headers=headers, stream=True)

            if response.status_code == 200:
                audio_bytes = b""
                for chunk in response.iter_content(chunk_size=CHUNK_SIZE):
                    if chunk:
                        audio_bytes += chunk

                audio_segment = AudioSegment.from_file(io.BytesIO(audio_bytes), format="mp3")

                # Oblicz długość segmentu na podstawie czasu rozpoczęcia i zakończenia
                segment_duration = end_time - start_time

                # Przygotowanie ciszy przed audio
                silence_before = AudioSegment.silent(duration=int(start_time * 1000))

                # Dodanie ciszy przed audio
                audio_segment = silence_before + audio_segment

                output_filename = os.path.join(output_folder, f"segment_{i}.mp3")
                audio_segment.export(output_filename, format="mp3")

                audio_segments.append(output_filename)

                # Obliczanie postępu
                progress = (i + 1) / total_subtitles * 100
                print(f"Progress: {progress:.2f}% ({i + 1}/{total_subtitles}) - Generating audio for: {text}")
            else:
                print(f"Failed to generate audio for: {text}")
        except Exception as e:
            print(f"Error: {e}")

    # Przesunięcie początkowe
    initial_shift = time_to_seconds(subtitles[0].start)

    # Łączenie segmentów przy użyciu FFmpeg z przesunięciem
    combined_output_path = os.path.splitext(path)[0] + '_calosc.mp3'

    # Tworzenie listy argumentów dla FFmpeg
    cmd = [
        "ffmpeg",
        "-i", f"concat:{'|'.join(audio_segments)}",
        "-af", f"aresample=async=1:first_pts=0",
        "-c:a", "mp3",
        combined_output_path
    ]
    subprocess.run(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
    print(f"Generated combined audio file: {combined_output_path}")

if __name__ == "__main__":
    arg_parser = argparse.ArgumentParser()
    arg_parser.add_argument("-p", "--path", help="subtitle file path", required=True)
    arg_parser.add_argument("-r", "--rate", help="speech rate (words per minute)", type=int, default=150)

    args = arg_parser.parse_args()

    generate_audio(path=args.path, rate=args.rate)

Możesz uściślić w którym momencie jest błąd? Co w ogóle mówi debuger?