# Media Transcriber - Audio, Video & Podcast to Text, SRT, VTT (`fabrikit/media-transcriber`) Actor

Transcribe audio, video and podcast feeds to text with timestamps, SRT and VTT. Bring your Groq/OpenAI key for full-length Whisper, or use the built-in small model. Language detection, English translation. From US$2 per 1,000 audio minutes.

- **URL**: https://apify.com/fabrikit/media-transcriber.md
- **Developed by:** [Fabrikit](https://apify.com/fabrikit) (community)
- **Categories:** AI, Automation, Developer tools
- **Stats:** 2 total users, 1 monthly users, 100.0% runs succeeded, 0 bookmarks
- **User rating**: No ratings yet

## Pricing

from $2.00 / 1,000 audio minute (your api key)s

This Actor is paid per event. You are not charged for the Apify platform usage, but only a fixed price for specific events.

Learn more: https://docs.apify.com/actors/running/actors-in-store.md#pay-per-event

## What's an Apify Actor?

An Actor is a serverless cloud program that runs on the Apify platform. It has two run modes.
In Batch mode, an Actor accepts a well-defined JSON input, performs an action which can take anything from a few seconds to a few hours,
and optionally produces a well-defined JSON output, datasets with results, or files in key-value store.
In Standby mode, an Actor provides a web server which can be used as a website, API, or an MCP server.

Apify vocabulary and the platform model are defined once, in the agent quickstart at https://apify.com/agents.md.

## How to integrate an Actor?

If asked about integration, you help developers integrate Actors into their projects.
You adapt to their stack and deliver integrations that are safe, well-documented, and production-ready.

Do not guess an integration path. Every one of them is in the agent quickstart at https://apify.com/agents.md: the Apify MCP server, Agent Skills with the Apify CLI, the JavaScript and Python clients, the REST API, and the account-free path for an agent with no human to sign in. It also carries the rule on stating cost before the first paid run.

For examples already wired to this Actor's own input schema, see the [API](#api) section below.

Each client library has reference documentation the quickstart does not restate: [JavaScript/TypeScript](https://docs.apify.com/api/client/js/docs.md) (`npm install apify-client`) and [Python](https://docs.apify.com/api/client/python/docs.md) (`pip install apify-client`).

# README

## Media Transcriber - audio, video & podcast to text, SRT and VTT

Turn audio and video files, Google Drive / Dropbox share links and podcast RSS feeds into accurate transcripts with timestamps, ready-made **SRT / VTT subtitles** and optional word-level timing. Language is detected automatically and any language can be translated to English.

Two engines, one output:

| Engine | What you need | Length per file | Price |
|---|---|---|---|
| **Your own API key** (Groq `whisper-large-v3-turbo` or OpenAI `whisper-1`) | A Groq key (free tier available) or an OpenAI key | Up to 12 hours | **US$0.002 per audio minute** (US$2 per 1,000 minutes) |
| **Built-in local model** (Whisper base, int8, CPU) | Nothing | First 10 minutes | US$0.05 per audio minute |

With your own key the actor does everything around the API call: it downloads the file, converts it with ffmpeg to a small mono MP3, **cuts long media at silences** (so no word is split) to respect the 25 MB API upload limit, calls the API in parallel, **merges the timestamps** and builds the SRT/VTT. The inference itself is billed by Groq/OpenAI directly to your account (Groq's turbo model costs roughly US$0.04 per audio hour at the time of writing).

Your key is an encrypted input field. It is kept in memory only and never logged, stored or written to the dataset.

### What you can feed it

- Direct links to `mp3, m4a, wav, ogg, opus, flac, aac, mp4, mov, webm, mkv...` files
- **Google Drive** and **Dropbox** public share links (rewritten to direct downloads, including Drive's "large file" confirmation page)
- **Podcast RSS / Atom feeds**: the newest N episodes are transcribed (`maxEpisodesPerFeed`)
- Files uploaded from your computer through the *Uploaded files* field

Not supported: YouTube and other streaming-site page URLs (they are not media files), password-protected or private links.

### Input example

```json
{
  "urls": [
    { "url": "https://example.com/interview.mp3" },
    { "url": "https://feeds.example.com/my-podcast.xml" }
  ],
  "apiKey": "gsk_...your Groq key...",
  "language": "auto",
  "translateToEnglish": false,
  "maxItems": 5,
  "maxEpisodesPerFeed": 3
}
```

Key options: `provider` (auto / groq / openai / local), `language` (`auto` or a code like `es`), `translateToEnglish`, `prompt` (vocabulary hint), `wordTimestamps`, `maxMinutesPerItem` (cap your spend), `chunkMinutes`, `includeSegments`, `includeSubtitles`, `saveFilesToKeyValueStore`.

### Output example

One dataset item per file or episode (failed files are listed with `status: "failed"` and an `error`, and are **not charged**):

```json
{
  "id": "96311eddd848",
  "sourceUrl": "https://example.com/interview.mp3",
  "title": "Interview with Ana",
  "status": "ok",
  "engine": "groq",
  "model": "whisper-large-v3-turbo",
  "task": "transcribe",
  "language": "es",
  "durationSecs": 1812.4,
  "billedMinutes": 31,
  "truncated": false,
  "text": "Hola y bienvenidos...",
  "segments": [{ "id": 0, "start": 0.0, "end": 4.2, "text": "Hola y bienvenidos..." }],
  "srt": "1\n00:00:00,000 --> 00:00:04,200\nHola y bienvenidos...\n",
  "vtt": "WEBVTT\n\n00:00:00.000 --> 00:00:04.200\nHola y bienvenidos...\n",
  "srtUrl": "https://api.apify.com/v2/key-value-stores/.../records/transcript-96311eddd848.srt"
}
```

The run's key-value store also holds `transcript-<id>.txt / .srt / .vtt` files and a `SUMMARY` record (minutes billed, engine, failures).

### Pricing

| Event | Price | When |
|---|---|---|
| `audio-minute` | US$0.002 | per audio minute (rounded up per file) with your Groq/OpenAI key |
| `audio-minute-local` | US$0.05 | per audio minute with the built-in model (max 10 min per file) |

Examples: a 1-hour podcast episode with your key = US$0.12. Ten 5-minute clips with the built-in model = US$2.50. Use `maxMinutesPerItem` and the run's maximum cost setting to cap spending; the actor stops cleanly when the limit is reached.

### FAQ

**Which engine should I pick?** Use your own Groq key for anything longer than a few minutes or when accuracy matters: large-v3-turbo is far more accurate than the small local model, and much faster. The built-in model is for quick tests and short clips without any account.

**Does it keep my files?** No. Downloads and converted chunks are deleted as soon as each file is processed. Only the transcript is stored in your run's dataset and key-value store.

**What about speakers (diarization)?** Not included; the Whisper APIs do not provide it.

**A link failed.** The error says why (private link, not a media file, too large, no audio track). For Google Drive, sharing must be "Anyone with the link". The actor tries a direct download first and only escalates to Apify proxies if the host blocks it.

**Accuracy of translation?** With Groq the translation endpoint uses `whisper-large-v3`; with OpenAI `whisper-1`. The built-in model can also translate, with lower quality.

### Use cases

Podcast show notes and search, meeting and interview transcripts, subtitles for videos, content repurposing, RAG over audio archives, accessibility, call recordings review, lecture notes.

### En español

Transcribe audio, video y podcasts (enlaces directos, Google Drive, Dropbox o un feed RSS) a texto con marcas de tiempo, subtítulos SRT/VTT y detección automática del idioma; también puede traducir al inglés.

- **Con tu propia API key** de Groq (hay capa gratuita) u OpenAI: archivos de hasta 12 horas, US$0,002 por minuto de audio. El actor descarga, convierte, corta en los silencios, une los tiempos y arma los subtítulos. La key es un campo secreto y nunca se guarda ni se registra.
- **Sin key**: modelo Whisper pequeño integrado, solo los primeros 10 minutos de cada archivo, US$0,05 por minuto.
- Los archivos descargados se borran al terminar. No se admiten enlaces de YouTube.

# Actor input Schema

## `urls` (type: `array`):

Direct links to mp3, mp4, m4a, wav, webm, ogg, flac, mov... files, Google Drive or Dropbox share links (public), or a podcast RSS feed URL (the latest episodes are transcribed). YouTube links are not supported.

## `files` (type: `array`):

Optional: upload audio/video files from your computer (stored in a key-value store of your account). Each uploaded file must be publicly reachable by URL, which is what the upload dialog provides.

## `apiKey` (type: `string`):

Bring your own key to transcribe full-length media with Whisper large-v3-turbo (Groq, free tier available) or whisper-1 (OpenAI). Keys starting with gsk\_ are used with Groq, sk- with OpenAI. The key is stored encrypted, kept in memory only and never logged or saved. Leave empty to use the built-in local model.

## `provider` (type: `string`):

Auto uses Groq/OpenAI when you provide a key (detected from its prefix) and the built-in local model otherwise.

## `language` (type: `string`):

"auto" detects the language. Or give a Whisper language code such as en, es, de, fr, pt, it, ja, zh. Setting it improves accuracy for short or noisy audio.

## `translateToEnglish` (type: `boolean`):

Return the English translation instead of the original-language transcript (Whisper translation). The detected source language is then not reported.

## `prompt` (type: `string`):

Optional names, jargon or style hint (up to ~200 words) that helps the API model spell rare words correctly. Ignored by the local model.

## `wordTimestamps` (type: `boolean`):

Also return a `words` array with start/end for every word. Only with a Groq/OpenAI key.

## `maxItems` (type: `integer`):

Upper limit of audio/video files processed in this run (podcast episodes count one each).

## `maxEpisodesPerFeed` (type: `integer`):

How many of the newest episodes to transcribe when a URL is an RSS feed.

## `maxMinutesPerItem` (type: `integer`):

Transcribe only the first N minutes of each file to cap your spend. 0 means the full file (local mode is always capped at 10 minutes).

## `chunkMinutes` (type: `integer`):

Long media is cut at silences into chunks of about this length before it is sent to the API (API upload limit is 25 MB per request).

## `localModel` (type: `string`):

Only for the no-key mode. Base is more accurate; tiny is a bit faster.

## `includeSegments` (type: `boolean`):

Add the `segments` array (start, end, text) to every dataset item.

## `includeSubtitles` (type: `boolean`):

Add `srt` and `vtt` strings to every dataset item.

## `saveFilesToKeyValueStore` (type: `boolean`):

Also store the transcript as downloadable .txt, .srt and .vtt files in the run's key-value store (links are returned as txtUrl, srtUrl, vttUrl).

## `maxFileSizeMb` (type: `integer`):

Files larger than this are skipped.

## `proxyConfiguration` (type: `object`):

Only used to download your URLs. The actor downloads directly first and escalates to Apify datacenter and then residential proxies only if the host blocks it.

## Actor input object example

```json
{
  "urls": [
    {
      "url": "https://upload.wikimedia.org/wikipedia/commons/7/7c/Marcus_Garvey%2C_speech%2C_1921.ogg"
    }
  ],
  "provider": "local",
  "language": "auto",
  "translateToEnglish": false,
  "wordTimestamps": false,
  "maxItems": 1,
  "maxEpisodesPerFeed": 3,
  "maxMinutesPerItem": 1,
  "chunkMinutes": 20,
  "localModel": "base",
  "includeSegments": true,
  "includeSubtitles": true,
  "saveFilesToKeyValueStore": true,
  "maxFileSizeMb": 1000,
  "proxyConfiguration": {
    "useApifyProxy": true
  }
}
```

# Actor output Schema

## `transcripts` (type: `string`):

One item per audio/video file (or podcast episode)

## `summary` (type: `string`):

Minutes billed, engine, failures

# API

You can run this Actor programmatically using our API. Below are code examples in JavaScript, Python, and CLI, as well as the OpenAPI specification and MCP server setup.

## JavaScript example

```javascript
import { ApifyClient } from 'apify-client';

// Initialize the ApifyClient with your Apify API token
// Replace the '<YOUR_API_TOKEN>' with your token
const client = new ApifyClient({
    token: '<YOUR_API_TOKEN>',
});

// Prepare Actor input
const input = {
    "urls": [
        {
            "url": "https://upload.wikimedia.org/wikipedia/commons/7/7c/Marcus_Garvey%2C_speech%2C_1921.ogg"
        }
    ],
    "provider": "local",
    "maxItems": 1,
    "maxMinutesPerItem": 1,
    "proxyConfiguration": {
        "useApifyProxy": true
    }
};

// Run the Actor and wait for it to finish
const run = await client.actor("fabrikit/media-transcriber").call(input);

// Fetch and print Actor results from the run's dataset (if any)
console.log('Results from dataset');
console.log(`💾 Check your data here: https://console.apify.com/storage/datasets/${run.defaultDatasetId}`);
const { items } = await client.dataset(run.defaultDatasetId).listItems();
items.forEach((item) => {
    console.dir(item);
});

// 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/js/docs

```

## Python example

```python
from apify_client import ApifyClient

# Initialize the ApifyClient with your Apify API token
# Replace '<YOUR_API_TOKEN>' with your token.
client = ApifyClient("<YOUR_API_TOKEN>")

# Prepare the Actor input
run_input = {
    "urls": [{ "url": "https://upload.wikimedia.org/wikipedia/commons/7/7c/Marcus_Garvey%2C_speech%2C_1921.ogg" }],
    "provider": "local",
    "maxItems": 1,
    "maxMinutesPerItem": 1,
    "proxyConfiguration": { "useApifyProxy": True },
}

# Run the Actor and wait for it to finish
run = client.actor("fabrikit/media-transcriber").call(run_input=run_input)

# Fetch and print Actor results from the run's dataset (if there are any)
print(f"💾 Check your data here: https://console.apify.com/storage/datasets/{run.default_dataset_id}")
for item in client.dataset(run.default_dataset_id).iterate_items():
    print(item)

# 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/python/docs/quick-start

```

## CLI example

```bash
echo '{
  "urls": [
    {
      "url": "https://upload.wikimedia.org/wikipedia/commons/7/7c/Marcus_Garvey%2C_speech%2C_1921.ogg"
    }
  ],
  "provider": "local",
  "maxItems": 1,
  "maxMinutesPerItem": 1,
  "proxyConfiguration": {
    "useApifyProxy": true
  }
}' |
apify call fabrikit/media-transcriber --silent --output-dataset

```

## MCP server setup

```json
{
    "mcpServers": {
        "apify": {
            "type": "http",
            "url": "https://mcp.apify.com/?tools=fetch-actor-details,fabrikit/media-transcriber"
        }
    }
}
```

The hosted server signs you in with OAuth on first connect, so no API token belongs in this config. Clients without OAuth support can send an `Authorization: Bearer <APIFY_API_TOKEN>` header instead, using a token from API & Integrations in Apify Console (https://console.apify.com/settings/integrations).

## OpenAPI specification

Download the OpenAPI definition: https://api.apify.com/v2/actors/e12doDVfpJgyiehAo/builds/rkM3YobCA4djX5CRZ/openapi.json