Start from Audio

Use this feature to turn any existing audio file — a voice recording, audiobook, podcast, or song — into a fully visualized AI-generated video. The system automatically transcribes your audio and lets you style and animate it visually.

Perfect for:

• 📚 Converting audiobooks into cinematic video books
• 🎵 Creating music videos from existing songs or instrumentals
• 🗣️ Visualizing voiceovers, spoken poetry, or podcast moments

Step-by-Step Workflow

Step 1: Upload Audio

Click Upload Audio and drag in your MP3 or WAV file.

Supported formats: .mp3 and .wav | Max size: 200MB. You’ll see your file name listed and a Change File option once uploaded.

Step 2: Select Language

After uploading, you'll see a preview player. Choose the language spoken in your audio:

• Automatic (Recommended) — Auto-detects language and accent
• Or manually select (e.g., English, Hindi, Tamil, Spanish)

Note: If the language is misidentified, you can go back and correct it before transcription.

Step 3: Transcribe

Click the red Transcribe button to convert your audio into timed script text.\

Credits used will be shown.

The system will extract dialogue, detect pauses, and structure it as a script for visual rendering.

Next Steps:

After transcription, you'll continue through the standard flow:

• Set Style → Choose visuals, camera tone, animation
• Finalize Cast → Match characters to voice
• Finalize Screenplay → Refine scenes if needed

🎯 Pro Tips

• For audiobook-to-video workflows: Use slow or medium scene pacing, and choose a thematic style that match the book’s tone. You can also custom train your own AI characters and AI styles for the video.
• For music videos: Let the AI detect lyrical structure. Choose dynamic visual styles with fast scene pacing for rhythm sync.
• Want full control? After transcription, you can switch to Screenplay mode to manually adjust scenes and characters.

🛠 Troubleshooting

• Upload failed? Check file format and size.
• Audio not transcribing well? Try cleaner audio or reduce background noise.
• Incorrect language? Select it manually before clicking Transcribe.

Still Need Help? Reach out to us at support@atlabs.ai — we’re happy to assist with book, voice, or music-based storytelling.

Updated on: 10/23/2025