Free AI Video Maker — Open Skill
Free AI Video Maker
Free AI Video Maker is a public Open Skill on Digen. Open it to see what this workflow creates.
Roleplay
4 1 7
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/1648313/2026-09-26/89gj9x4m.png
# Free AI Video Maker Turn a user's idea into a short AI video (with voiceover and background music) using the platform's **built-in free generators**. No external API keys, accounts, or credits are needed — everything runs on the platform's own `generate_image` / `generate_video` / `generate_tts` / `generate_music` / `mix_clips` tools. > Honest framing: "free / no credit" here means the built-in platform generators, which need no third-party API key or payment. The skill cannot call arbitrary third-party free APIs (the toolset is fixed). If the user insists on a specific external API, politely explain this limitation and offer the built-in path instead. ## When to use - User wants an AI video made from a one-line idea, a short script, or an uploaded photo/audio. - User wants a narrated / talking-style video, a product pitch, or a short cinematic clip. - User asks for "free AI video", "make a video with no credit", "generate a video + voiceover + music". ## Steps ### 1. Clarify the brief (one round, keep it light) Ask only what's missing and essential: - **Idea or script**: do they have a script, or should you write one from a one-line idea? - **Orientation**: vertical (9:16, for TikTok/Reels/Shorts) or horizontal (16:9)? Default to vertical for short-form. - **Length**: default ~10–15 seconds for a short clip; longer if they ask. - **Voiceover**: do they want a narrator voice, and in what language/tone? - **Reference material**: any uploaded photo (to feature as the subject) or audio (to drive the video)? ### 2. Write the script (if none provided) From the idea, write a short script (2–4 short sentences for a ~10–15s clip). Keep it punchy. If the user gave a script, use it as-is. ### 3. Generate the voiceover (optional but recommended) Use `generate_tts` with `model='tts.design'` and a `voice_instruction` (in English) describing the narrator's voice (e.g. "warm, clear female narrator, calm and friendly"). Pass the script text as `text`. Set `name` and `entity_type` (e.g. `name="voiceover"`, `entity_type="voiceovers"`). ### 4. Generate the video clip(s) Pick the channel by what's available: - **No reference image** (pure text idea): `generate_video` with `model='t2v'`, no `input_urls`. Describe the scene in the `prompt`. - **Uploaded photo as the subject**: `generate_video` with `model='i2v'` and the photo in `input_urls` (first frame). Describe the motion in the `prompt`. - **Reference image(s) for consistent identity**: `generate_video` with `model='ref2v'` and the reference image(s) in `input_urls`. - **Audio-driven scene**: if the user uploaded audio, use `model='ref2v'` (or `i2v`) plus `audio_url` (and optionally `audio_start_time`). Set `orientation` to match the brief (portrait for 9:16, landscape for 16:9), `resolution` default `720P`, and `duration` to roughly match the voiceover length. Give each clip a clear `name` and `entity_type` (e.g. `name="clip-1"`, `entity_type="storyboard_videos"`). > Video generation does **not** support NSFW/adult content on any channel. If the user requests adult content, politely decline and offer a safe alternative. ### 5. Generate background music (optional) Use `generate_music` with a `prompt` describing the mood (e.g. "soft ambient electronic background, uplifting, no vocals"). Keep it instrumental so it doesn't clash with the voiceover. Set `name` and `entity_type` (e.g. `name="bgm"`, `entity_type="music"`). ### 6. Mix everything into the final video Use `mix_clips` to combine the video clip(s) with the voiceover and music: - `clips`: one clip per video segment, with `video.url` = the generated clip asset, `audio.url` = the voiceover, and `audios` = the music track (with a lower `volume`, e.g. `0.3`, so it sits under the voice). - Set `orientation` to match the clips (portrait/landscape), `name` = final title, `entity_type="final_videos"`. ### 7. Deliver Show the final video to the user. Summarize what was made (script, voiceover, music, clip count, orientation, length)