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

# Introduction

> AI-powered lyrics transcription and alignment for music — specialising in Indian languages

lyrcs.ai is a REST API that takes an audio file and returns machine-readable lyrics: the original script, a phonetic transliteration, an English translation, and time-synced LRC and SRT files ready for delivery to streaming platforms, karaoke systems, or rights management databases.

<CardGroup cols={2}>
  <Card title="48 languages" icon="globe">
    30 Indian languages including all 22 scheduled languages, plus 18 global languages. Indian language audio has additional prompt layers and correction dictionaries for higher accuracy.
  </Card>

  <Card title="Three output modes" icon="sliders">
    Transcript-only for metadata workflows (\~60s), full pipeline with LRC/SRT (\~90s), or full pipeline with an artist review gate before delivery.
  </Card>

  <Card title="Webhook delivery" icon="webhook">
    Five event types — `job.complete`, `job.awaiting_review`, `job.degraded`, `job.failed`, `batch.complete` — with HMAC-SHA256 signature verification and automatic retry.
  </Card>

  <Card title="Batch processing" icon="layer-group">
    Submit up to 20 jobs in a single request. A single `batch.complete` webhook fires when all jobs are done.
  </Card>
</CardGroup>

## Who it's for

lyrcs.ai is built for organisations that handle music at scale:

* **Music distributors** — deliver time-synced lyrics to DSPs alongside audio
* **Music labels** — process catalog at scale across multiple artists and languages, with optional artist approval workflows
* **Regional streaming platforms** — power lyric display and karaoke features without building transcription infrastructure
* **Rights management companies** — generate structured lyric metadata for sync licensing and catalog databases
* **Lyrics and metadata providers** — supply accurate, time-synced lyrics data for Indian language catalogs at scale
* **Karaoke providers** — produce LRC files from raw audio

## Three API modes

Every transcription job runs in one of three modes. Choose based on what you need delivered and how fast.

| Mode              | Parameters                   | Output                                     | Time                |
| ----------------- | ---------------------------- | ------------------------------------------ | ------------------- |
| Transcript only   | `align=false`                | Transcript + transliteration + translation | \~60s               |
| Full pipeline     | `align=true` (default)       | All of above + LRC + SRT                   | \~90s               |
| Reviewed delivery | `align=true` + `review=true` | All of above, held until artist approves   | \~90s + review time |

See the [Three API Modes guide](/guides/three-api-modes) for a full breakdown.

## Next steps

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="/getting-started/quickstart">
    Submit your first job and download an LRC file in under 5 minutes.
  </Card>

  <Card title="Authentication" icon="key" href="/getting-started/authentication">
    Learn how API keys work and how to request access.
  </Card>
</CardGroup>