voice_wordpress_article.py

from openai import OpenAI
import requests
import re
import os
from pydub import AudioSegment
from tqdm import tqdm
    
def getLatest():
    data = requests.get("https://volksverpetzer.de/wp-json/wp/v2/posts/", headers={'User-Agent': 'Voice'})
    #replace all html tags
    content = data.json()
    article = content[0]["content"]["rendered"]
    slug = content[0]["slug"]
    title = content[0]["title"]["rendered"]
    article = re.sub(r'<[^>]*>', '', article)
    return article, slug, title
    
def voiceArticle(article):
    client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])
    from langchain.text_splitter import RecursiveCharacterTextSplitter
    text_splitter = RecursiveCharacterTextSplitter(
        chunk_size=4000,
        chunk_overlap=0,
        length_function=len
    )

    chunks = text_splitter.create_documents([article])
    for i, chunk in tqdm(enumerate(chunks)):
        with client.audio.speech.with_streaming_response.create(
            model="tts-1-hd",
            voice="shimmer",
            input=chunk.page_content
        ) as response:
            response.stream_to_file("output{}.mp3".format(i))

def combineVoices():
    files = [f for f in os.listdir() if f.endswith(".mp3")]
    print(files)
    #exclude output.mp3
    files = [f for f in files if f != "output.mp3"]
    # sort files by name 
    files.sort(key=lambda x: int(x[6:-4]))
    combined = AudioSegment.empty()
    for file in files:
        combined += AudioSegment.from_mp3(file)
    combined.export("output.mp3", format="mp3")
    for file in files:
        os.remove(file)
    
def uploadVoice(slug, title):
    # File path and access token
    file_path = 'output.mp3'
    access_token = os.environ["SOUNDCLOUD_ACCESS_TOKEN"]

    headers = {
        'accept': 'application/json; charset=utf-8',
        'Authorization': 'OAuth {}'.format(access_token),
    }
    print(title)
    files = {
        'track[title]': (None, title),
        'track[asset_data]': open(file_path, 'rb'),
        "track[sharing]": (None, "public"),
        "track[permalink]": (None, slug),
    }
    
    print(files)    

    response = requests.post('https://api.soundcloud.com/tracks', headers=headers, files=files)

    track_info = response.json()

    if response.status_code == 201:
        print("Track uploaded successfully!")
        print("Track URL:", track_info['permalink_url'])
    else:
        print("Failed to upload track:", response.content)
        
def main():
    article, slug, title = getLatest()
    voiceArticle(article)
    combineVoices()
    uploadVoice(slug, title)
    
if __name__ == "__main__":
    main()