---
title: Audio Preview
url: https://docs.kaltura.com/docs/conversational-agents/avatar-videos/audio-preview
---

## Audio Preview

Preview the text-to-speech narration for a specific scene before generating the full video:

```bash
# Returns audio/mpeg binary
curl -s -X POST "$AVATAR_API/video/previewAudio" \
  -H "Authorization: Bearer $KS" \
  -H "Content-Type: application/json" \
  -d "{ \"id\": \"$VIDEO_ID\", \"sceneId\": 0 }" \
  --output scene_preview.mp3
```

**Request fields:**

| Field | Type | Required | Description |
|-------|------|----------|-------------|
| `id` | string | yes | Video project ID |
| `sceneId` | number | yes | Scene index (0-based) |

The scene requires narration text to be populated — returns `SCENE_EMPTY_NARRATION` if the text is missing.

Use `previewAudioStream` for streaming playback instead of downloading the full file:

```bash
curl -s -X POST "$AVATAR_API/video/previewAudioStream" \
  -H "Authorization: Bearer $KS" \
  -H "Content-Type: application/json" \
  -d "{ \"id\": \"$VIDEO_ID\", \"sceneId\": 0 }" \
  --output scene_stream.mp3
```