> ## Documentation Index
> Fetch the complete documentation index at: https://docs.neolli.jocoding.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Transcription

> Convert spoken audio to timestamped captions with speaker identification.

## Overview

Transcription converts the spoken audio in your video into timestamped captions in the source language. Powered by **ElevenLabs Scribe v2**, it produces word-level timing and speaker diarization — the essential foundation for translation and dubbing.

## Starting a transcription

1. Open your video from the dashboard
2. Click **Transcribe**
3. Select the spoken language, or choose **Auto-detect**
4. Click **Start**

<Tip>
  **Auto-detect** works well for most common languages. Select a specific language if you know it — this improves accuracy, especially for less common languages like Dutch or Chinese.
</Tip>

## Supported languages

Neolli supports transcription in **10 languages** (plus auto-detect):

| Flag | Language   | Code  |
| ---- | ---------- | ----- |
| 🇺🇸 | English    | `eng` |
| 🇪🇸 | Spanish    | `spa` |
| 🇫🇷 | French     | `fra` |
| 🇩🇪 | German     | `deu` |
| 🇮🇹 | Italian    | `ita` |
| 🇧🇷 | Portuguese | `por` |
| 🇯🇵 | Japanese   | `jpn` |
| 🇰🇷 | Korean     | `kor` |
| 🇨🇳 | Chinese    | `zho` |
| 🇳🇱 | Dutch      | `nld` |

For the full capabilities matrix across all features, see [Supported Languages](/workflows/supported-languages).

## Features

* **Speaker diarization** — Automatically identifies and labels different speakers
* **Word-level timing** — Each word gets its own precise timestamp for accurate syncing
* **Auto-detect** — Identifies the spoken language automatically for common languages

## Processing time

| Video length | Estimated time         | Mode         |
| ------------ | ---------------------- | ------------ |
| Under 30 min | 1–3 minutes            | Synchronous  |
| Over 30 min  | Proportional to length | Asynchronous |

You can close the browser while a job is running — it continues in the background. The dashboard shows job progress in real time.

## File requirements

* **Max file size**: 3 GB
* **Supported formats**: MP4, MOV, MKV, AVI, WebM, and most common video/audio formats
* **Audio**: Must contain a detectable audio track with speech

<Warning>
  Transcription accuracy depends heavily on audio quality. Background noise, overlapping speakers, heavy music, and low recording quality will reduce accuracy. See [Audio Quality Tips](/troubleshooting/audio-quality) for guidance.
</Warning>

## After transcription

Once complete, you can:

* Review and edit captions in the [caption editor](/workflows/caption-editor)
* Add target languages for [translation](/workflows/translation) or [dubbing](/workflows/dubbing)
* [Export](/workflows/exporting) the source captions as SRT

## Credit cost

Transcription is charged at **58 credits per minute** of audio. See [Credit Costs](/billing/credit-costs) for a detailed breakdown.