Images plus a script become a narrated video with captions and music in one ReelSDK call to POST /v1/slideshow. A 23-second video costs about 2.7 cents: $0.03 per video minute plus $0.05 per 1,000 characters of speech.

Turn images plus a script into a narrated video

Send image URLs and the script as voiceover text, and ReelSDK reads it in one of 40 voices, times the photos to it and returns the MP4 in under 30 seconds.

Use casesUpdated

4 photos of Rome and Venice timed to the voiceover, with bold captions, ready in under 30 seconds.Input: first photocaptions.srttranscript.json

The request

POST /v1/slideshow

Send images and a script; get back a video narrated in one of 40 voices, with captions and music under the voice.

terminalTry it
curl "https://api.reelsdk.com/v1/slideshow?wait=300" \
  -H "Authorization: Bearer $REELSDK_KEY" \
  --json '{
    "images": [
      "https://api.reelsdk.com/f/lib/in/trips/rome-3.jpg",
      "https://api.reelsdk.com/f/lib/in/trips/rome-1.jpg",
      "https://api.reelsdk.com/f/lib/in/trips/rome-2.jpg",
      "https://api.reelsdk.com/f/lib/in/trips/venice.jpg"
    ],
    "size": "9:16",
    "voiceover": {
      "text": "Three days, two cities, one train ride. Start in Rome at the Colosseum, an hour before the crowds arrive. Walk the old arches while the stone is still cool. Then throw a coin into the Trevi Fountain, so you come back one day. On day three, take the fast train north, and end the trip on a gondola in Venice.",
      "voice": "apollo"
    },
    "music": {
      "url": "https://api.reelsdk.com/f/lib/music/cinematic/spring-on-the-horizon.mp3",
      "volume": 0.18
    },
    "captions": "bold"
  }'
request.mjsTry it
const res = await fetch("https://api.reelsdk.com/v1/slideshow?wait=300", {
  method: "POST",
  headers: { Authorization: `Bearer ${process.env.REELSDK_KEY}`, "Content-Type": "application/json" },
  body: JSON.stringify({
    "images": [
      "https://api.reelsdk.com/f/lib/in/trips/rome-3.jpg",
      "https://api.reelsdk.com/f/lib/in/trips/rome-1.jpg",
      "https://api.reelsdk.com/f/lib/in/trips/rome-2.jpg",
      "https://api.reelsdk.com/f/lib/in/trips/venice.jpg"
    ],
    "size": "9:16",
    "voiceover": {
      "text": "Three days, two cities, one train ride. Start in Rome at the Colosseum, an hour before the crowds arrive. Walk the old arches while the stone is still cool. Then throw a coin into the Trevi Fountain, so you come back one day. On day three, take the fast train north, and end the trip on a gondola in Venice.",
      "voice": "apollo"
    },
    "music": {
      "url": "https://api.reelsdk.com/f/lib/music/cinematic/spring-on-the-horizon.mp3",
      "volume": 0.18
    },
    "captions": "bold"
  }),
});
const { job } = await res.json();
response.json
{
  "job": {
    "id": "job_87gbensmzk2i1rgrl4ny",
    "op": "slideshow",
    "status": "succeeded",
    "stage": null,
    "created_at": "2026-11-07T01:31:37.401Z",
    "finished_at": "2026-11-07T01:32:06.516Z",
    "outputs": {
      "video": "https://api.reelsdk.com/f/out/job_87gbensmzk2i1rgrl4ny/video.mp4?s=...",
      "transcript": "https://api.reelsdk.com/f/out/job_87gbensmzk2i1rgrl4ny/transcript.json?s=...",
      "srt": "https://api.reelsdk.com/f/out/job_87gbensmzk2i1rgrl4ny/captions.srt?s=...",
      "vtt": "https://api.reelsdk.com/f/out/job_87gbensmzk2i1rgrl4ny/captions.vtt?s=...",
      "ass": "https://api.reelsdk.com/f/out/job_87gbensmzk2i1rgrl4ny/captions.ass?s=..."
    },
    "result": null,
    "duration_s": 23.166,
    "error": null,
    "usage": {
      "output_seconds": 23.166,
      "audio_seconds": 22.656,
      "tts_characters": 307,
      "images": 0
    },
    "poll_url": "https://api.reelsdk.com/v1/jobs/job_87gbensmzk2i1rgrl4ny"
  }
}

Timing follows the narration

When you send a voiceover and leave out seconds_per_image, the photos divide the narration length evenly. Here the voice ran just under 23 seconds and the video half a second longer, so each of the 4 photos stayed on screen for almost 6 seconds.

You can still pin the timing on individual slides when you need to. Give a photo its own seconds value and it holds that exact duration, while the remaining photos split whatever time is left over, with seconds_per_image supporting anywhere from 0.5 to 60.

At a glance

Three ways to send the voiceover

Three ways to send the voiceover
FormIn the requestBilled
A recording"voiceover": "https://.../voice.mp3"Output minutes only
A recording with a volume{ "url": "https://.../voice.mp3", "volume": 0.8 }Output minutes only
Text to speech, this run{ "text": "...", "voice": "apollo" }Plus $0.05 per 1,000 characters

Sources: ReelSDK OpenAPI spec, ReelSDK pricing, checked November 2026.

Volume runs from 0 to 2. Text takes up to 10,000 characters in one of 40 English voices.

Captions come from the voice

Enabling captions requires a voiceover, since the audio track gets transcribed with whisper-large-v3-turbo to time each word to the speech. You do not need to build a subtitle file yourself or calculate timestamps before making the call.

This request uses the bold preset, which sets the text in Montserrat, uppercase, 3 words per group in the middle of the frame, and highlights the spoken word as it plays. You can also adjust max_words from 1 to 8 and size from 0.5 to 2.

Music under the voice

Background audio is looped or trimmed to match the video length and faded out over the last 1 s. It also gets ducked under the voiceover, so the track drops whenever the narration is speaking and comes back up in the pauses.

This run sets volume to 0.18 on a scale that runs from 0 to 2. You can pick from the built-in library of 16 CC0 tracks in 8 moods by HoliznaCC0, or pass any MP3 URL you host yourself. Full rates are on pricing and every request field is listed in the slideshow docs.

What it costs

What this video cost

Video, 23.2 s at $0.03 per minute
$0.0116
Voiceover, 307 characters at $0.05 per 1,000
$0.0154
Total
$0.0269

Billed per second of output. A failed job costs nothing.

Limits

What one request takes

Images
1 to 100 per video
Seconds per image
0.5 to 60
Voiceover text
Up to 10,000 characters
Input files
500 MB each, 1 GB per job
Media length
Up to 15 minutes for captions and slideshows
Job run time
Up to 10 minutes
Concurrent jobs
3 per key
Jobs per day
500 per key
Output retention
7 days

How it works

From request to file

  1. Pick the images

    Pass 1 to 100 public image URLs in the order they should play.

  2. Write the script

    Put up to 10,000 characters of plain text in voiceover.text.

  3. Choose a voice and music

    Pick one of the 40 English voices, add a music URL from the 16 CC0 tracks or your own host, and set volume between 0 and 2.

  4. Send POST /v1/slideshow

    Call POST /v1/slideshow with ?wait=300 so the rendered video, captions.srt and transcript.json return in the same response.

FAQ

Questions, answered

What did this narrated video cost?

About 2.7 cents: $0.03 per minute of video plus $0.05 per 1,000 characters of speech, and the speech is more than half of that. The $0.50 signup credit covers 2 videos like this one, because starting a job holds the most it can cost, about 47 cents, until it finishes.

Which voices are there?

There are 40 English text to speech voices available. This run uses apollo, and luna is the default if you do not pick one.

Can I use my own recording?

Yes, pass an MP3 URL in the voiceover field. You are only billed for the output minutes with no speech charge.

Can the narration be in another language?

The 40 built-in voices are English only. For another language, pass your own audio recording and set the language field so the captions get transcribed in that language.

How long can the script be?

A script can be up to 10,000 characters long. The finished video can run up to 15 minutes, and a render job can run for up to 10 minutes.

Ready to build?

Ship video from your app today

Your first $0.50 is free, and one key covers every endpoint.

Moonlit lake with a lit cabin under the mountains