
Turn PDFs into audiobooks with the AI you already pay for
PDF TO MP3 turns PDF documents into audiobooks. It can be used two ways: connect your existing Codex or Claude subscription via an MCP server to handle document processing (so PDF TO MP3 itself costs $0, with speech generation billed directly by OpenAI or Google), or use fully hosted plans where PDF TO MP3 handles processing and billing. Documents can be listened to on the web, in iPhone/Android apps, or downloaded as MP3s.
Connect Codex or Claude by adding the PDF TO MP3 MCP server, so document processing runs through the AI subscription you already pay for at no extra cost from PDF TO MP3.
Add an OpenAI or Gemini API key that pays for speech generation directly, billed per usage by OpenAI or Google rather than by PDF TO MP3.
Alternatively, upload documents and let PDF TO MP3 handle all processing and usage billing, with credits covering processing, audio creation, questions, and live conversations.
Ask live questions about your documents and use the Ask and Explain buttons; credits are consumed for questions and live conversations.
Upload a document in one language and generate the audio in another to hear a translation.
Download a full MP3 for the whole document or per-chapter MP3s, and listen on the web or in the iPhone and Android apps.
Paid plans include premium text-to-speech and multi-speaker audio creation.
Sign in, choose Codex or Claude, and in Codex open Settings → Plugins → Add MCP server with Streamable HTTP, paste the provided server address, then authenticate to connect your PDF TO MP3 account.
Choose OpenAI or Gemini to make your speech and add the corresponding API key, which pays for speech generation only. Enable billing in the Gemini console for improved reliability.
Attach a PDF in your AI chat and ask it to use the PDF TO MP3 MCP to turn the document into an audiobook. Your subscription covers document processing via MCP, and your API key covers speech generation.
Two cost models: (1) Use your existing AI subscription — $0 to PDF TO MP3; document processing runs through your Codex or Claude subscription and speech generation is billed directly by OpenAI or Google. (2) Hosted plans where PDF TO MP3 handles processing and billing: Free ($0, 3,000 words/month), Weekly ($4.99, 150 credits/week), Monthly ($12.99/month, 600 credits/month), Yearly ($89.99/year, 60
$0
Document processing via MCP covered by your existing Codex or Claude subscription; speech generation billed per usage directly by OpenAI or Google via your API key.
$0
Up to 3,000 words per month; create audio from the original text with one speaker.
$4.99
150 credits per week; full access to all features.
$12.99 per month
600 credits per month (3,000 words per month); upload documents, create audio, download MP3s, automatic translation, iPhone and Android app, Ask live questions, premium TTS and multi-speaker.
$89.99
600 credits per month on a yearly billing cycle; same full feature set as other paid plans.
The free plan includes up to 3,000 words per month. Weekly plans get 150 credits per week, and monthly and yearly plans get 600 credits per month. Credits are used for document processing, audio creation, asking questions, and live conversations.
All paid plans have the same features; only the payment cycle and number of credits differ. The free plan is different and only lets you create audio from the original text.
Yes. You can upload a document and later select which chapters you want to create audio for.
Yes. You can download a full MP3 for the whole document, and MP3s for each chapter.
Yes. Every paid plan gets full access to all features, including turning documents into audio, Ask and Explain buttons, and the iPhone and Android apps. The free plan only creates audio from the original text with one speaker.
Yes. When you cancel, your plan stays active until the end of your current billing period.
On the web or in the iPhone and Android apps. You can also download your audio as an MP3 and play it anywhere.
English, Spanish, French, German, Italian, Portuguese, Chinese, Japanese, Hindi, Korean, and Russian.