# Podcast Transcript API

> Podcast Transcript API is a paid MCP server for files and documents. Podcast transcripts from RSS, Apple Podcasts or audio URLs, with a hard spending cap. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Paid
- **Platforms:** API, Web
- **Website:** https://apify.com/catalyst-labs/podcast-transcript-api
- **Repository:** https://github.com/catalyst-labs-dev/catalyst-labs-mcp/tree/HEAD/servers/podcast-transcript-api
- **MCP Registry name:** io.github.catalyst-labs-dev/podcast-transcript-api
- **Categories:** files and documents, news and media
- **Listing updated:** 2026-11-15

## About

The cheapest path comes first: the tool checks whether a show publishes its own transcript through the Podcasting 2.0 tag and, if so, returns it for a flat $0.01 per episode. Only when none exists does it transcribe the audio with Whisper large-v3 (with a turbo fallback) at $0.009 per audio minute. Each record says whether the text came from the publisher or from speech-to-text.

The author's own spot check, run on 18 October 2026 and described as not an independent audit, found about 2.4 to 3.0 percent word error against publisher transcripts for English episodes, and 10.2 percent for Spanish and 16.9 percent for German. Speech-to-text output has no speaker labels, while publisher transcripts keep theirs.

Budget controls are checked before each episode, using maxAudioMinutes and maxTotalChargeUsd, so a run cannot overspend, and failed items are not charged. Output includes plain text, LLM-ready text, timestamped segments, SRT, VTT and Markdown. Spotify-exclusive and Apple-subscription shows have no public audio and are rejected without charge. It is hosted on Apify.

## Features

- **Publisher transcript first:** Uses a show's own transcript at $0.01 before paying for speech-to-text.
- **Whisper fallback:** Transcribes audio with Whisper large-v3 when no publisher transcript exists.
- **Hard budget caps:** Audio-minute and dollar limits are checked before each episode.
- **Many inputs:** Accepts RSS feeds, Apple Podcasts links, episode pages or audio URLs.
- **Several outputs:** Returns text, segments, SRT, VTT, Markdown and LLM-ready text.
- **Free failures:** Failed or rejected items are never charged.

## Good for

- Pull the latest episode transcript from a podcast RSS feed
- Feed podcast text into a research or summarising agent
- Make subtitles for an audio file in SRT or VTT
- Cap the cost of transcribing a backlog of episodes

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http podcast-transcript-api 'https://mcp.apify.com/?tools=catalyst-labs/podcast-transcript-api'
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://mcp.apify.com/?tools=catalyst-labs/podcast-transcript-api
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "podcast-transcript-api": {
      "url": "https://mcp.apify.com/?tools=catalyst-labs/podcast-transcript-api"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "podcast-transcript-api": {
      "type": "http",
      "url": "https://mcp.apify.com/?tools=catalyst-labs/podcast-transcript-api"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "podcast-transcript-api": {
      "serverUrl": "https://mcp.apify.com/?tools=catalyst-labs/podcast-transcript-api"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "podcast-transcript-api": {
      "type": "http",
      "url": "https://mcp.apify.com/?tools=catalyst-labs/podcast-transcript-api"
    }
  }
}
```

## Tools

- `catalyst-labs--podcast-transcript-api`: Starts a podcast transcript run from RSS, Apple Podcasts, episode or audio URLs and returns transcripts or a run ID to poll.

## FAQ

### How much does a transcript cost?

$0.01 when the show publishes its own, otherwise $0.009 per audio minute of speech-to-text.

### Can it overspend?

No. The budget is checked before each episode, and maxTotalChargeUsd sets a hard cap.

### Does it label speakers?

Publisher transcripts keep their speaker labels; speech-to-text does not.

### Does it work with Spotify-exclusive shows?

No, those have no public audio and are rejected without charge; pass the RSS feed instead.

### Where does it run?

On Apify, through a hosted remote MCP endpoint.

### Is Podcast Transcript API free?

No. Podcast Transcript API is a paid product.

### How do I add Podcast Transcript API to Claude Code?

Run this command in your terminal: claude mcp add --transport http podcast-transcript-api 'https://mcp.apify.com/?tools=catalyst-labs/podcast-transcript-api'

### What are the best alternatives to Podcast Transcript API?

The closest alternatives to Podcast Transcript API are ASTra, CARTESIAN ECU, Invoices, SigPen and ConvertAudioToText.

## Alternatives

- [ASTra](https://ai-powered.com/mcp/astra): ASTra is a freemium MCP server for files and documents.
- [CARTESIAN ECU](https://ai-powered.com/mcp/cartesian-ecu): CARTESIAN ECU is a paid MCP server for files and documents.
- [Invoices](https://ai-powered.com/mcp/invoices): Invoices is a paid MCP server for files and documents.
- [SigPen](https://ai-powered.com/mcp/sigpen): SigPen is a freemium MCP server for files and documents.
- [ConvertAudioToText](https://ai-powered.com/mcp/convertaudiototext): ConvertAudioToText is a freemium MCP server for design and media.
- [OpenNash Podcasts](https://ai-powered.com/mcp/opennash-podcasts): OpenNash Podcasts is a free MCP server for news and media.
- [WhoScribe](https://ai-powered.com/mcp/whoscribe): WhoScribe is a freemium MCP server for design and media.
- [filex](https://ai-powered.com/mcp/filex): filex is an open-source MCP server for files and documents.

---
Source: https://ai-powered.com/mcp/podcast-transcript-api
