--- name: filmscene-sound description: Re-voice the scene's dialogue in the person's characters' saved voices — isolate the voices, change each speaker span into the voice of the character who plays that role, lay the spans back at their times on a track exactly as long as the generation, and add the film's own room tone. --- # The actors' performance in your characters' voices FlashBoards user skill, version 1.2.3 (2026-10-13). On the board, read these guides with `list_skills` and `get_skill` when the assistant exposes them; otherwise use the hosted index through an available read-only retrieval tool. Use that runtime’s actual model discovery, current page context and guarded board actions for execution. Keep native readiness, approval and cost gates; do not claim a guide was read when retrieval is unavailable. Read related guides with `get_skill({skillId:""})` or the [hosted index](https://pub-78da82d51ed14147ac554d38094b0c0b.r2.dev/user-skills/1.2.3/index.html). Reading a guide does not start or authorize a paid generation. Read `filmscene` for the shared rules. > **Room tone first from the film.** The scene's own line-free gaps match it best. When the film has none, or an effect is missing, use ElevenLabs Sound Effects (`generate_audio` with `model: "elevenlabs-sound-effects-v2"`, `loop: true` for a bed) and level it down: a generated ambience reads louder than the film's own tone. > **Entry level:** with Turbo parts, change each speaker's voice in the finished part (`reels-lipsync-turbo`, step 3). ## Voices per character Each speaker span goes into the voice of the character who plays that role: {voice A} for {character A}, {voice B} for {character B} — the person's saved voices (`list_audio_voices`). A character without a voice → ask the person (or clone or pick one with them, `reels-voice`); never use demo voices. 1. **The part's audio**: `cut_audio_span` of the scene's track (`extract_video_audio` of the clip) for the part. 2. **Only the voices**: the ElevenLabs Audio Isolation model (`run_skill_model`, its id from `list_generation_models`). 3. **Speaker spans**: split by the speakers of `transcribe_media` (boundaries in the gaps between lines); label each span with its film role → its character; `cut_audio_span` each. 4. **Voice change**: `change_voice_audio` per span with that character's `voiceId`, `stability: 0.7`, `similarityBoost: 0.95`, `audioSource: {source: , startSeconds: 0, endSeconds: }`. Length and word times stay the actor's. Lower stability garbled the end of a mumbled line; 0.7 fixed it. 5. **Lay the spans back** at their offsets on one track exactly the generation's length: `mix_audio_tracks` with `durationSeconds` = the generation length and each changed span at its `offsetSeconds` → Audio 1. 6. **Check** with `transcribe_media`: the words and their times must match the original. Mumbled words vary; compare with the transcript of the isolated original before redoing. The actors' isolated voice is a working file — never published. ## Room tone - No clean background track exists: cut the scene's **gaps with no lines** (`cut_audio_span` where the transcript has no words) and join them (`concat_audio_tracks`) into one loop as long as the scene. - Level: as in the film, about 13 dB under the speech (`gain` about 0.22 in `mix_audio_tracks`), measured once and kept for every part. - Each part takes the loop from its own offset, so the parts join without a seam. A generated ambience (Sound Effects with `loop: true`) is the fallback when the film has no clean gaps: it reads louder, so set its level by ear against the film. ## Execute through MCP or on the board Use the person's FlashBoards account; never ask for a personal provider key or run the source's local scripts. Discover `get_account`, `list_generation_models` and the tools of this client first: a published guide does not prove a feature is deployed. **Headless, no open board.** Quote with `estimate_generation_cost` (skill models: `estimate_skill_model`), then run `generate_image`, `generate_video`, `generate_audio` or `run_skill_model` with one stable `generationAttemptId` per logical attempt. Pass `maxCostUsd` only when the current operation and selected model support it; otherwise use the quoted cost and the person's explicit approval with the supported controls. Never silently ignore an unsupported spending cap; make one `get_generation` check after `retryAfterSeconds`. Pass the person's own results by `generationId`: images in `referenceGenerationIds`, `firstFrameUrl` and `endFrameUrl`, audio in `referenceAudioUrls` or `targetAudioUrl`, videos in `referenceVideoGenerationIds`. They are read on the server; signed result links do not work as public URLs. Every `generationId` a tool returns stays valid: a later read of the same result gives a different string, and both work. Film frames and gallery pictures go in by `previewHandle`: `referencePreviewHandles` of the generation tools, a `{previewHandle}` source of `arrange_images`, `crop_image` and `pad_image`, and `describe_media`. New files come from `show_media_upload_picker` as receipts. The person's character comes in as its sheet, by `generationId` or as an upload; there is no search by name. If the person names a character without the sheet, ask for it or make one with `character-sheet`; do not scan recent results for it. Free server steps: `trim_video`, `concat_videos`, `conform_video`, `replace_video_audio`, `extract_video_frame`, `extract_video_audio`, `detect_video_cuts`, `crop_image`, `pad_image`, `arrange_images`, `cut_audio_span`, `concat_audio_tracks`, `mix_audio_tracks`, `fit_audio_duration`, `add_audio_reverb`. Check every result with `describe_media` (a question about an image; `model`, `mode` and `sourceRange` only for a video) and `transcribe_media` for speech. Both checks are paid: `describe_media` costs a fraction of a cent for an image and about 2–4 ¢ for a short video (more for a long one), `transcribe_media` under 1 ¢ a minute — count the checks you plan in the price you name before the first paid step. `describe_media` is not reliable about sound: it misses effects mixed under music and can hear voices or instruments that are not there. Check speech with `transcribe_media` and ask the person to listen for effects and music. Never repay an uncertain submission; reuse an attempt ID only for the identical request. Make each new `requestId` or `generationAttemptId` unique, for example the task, the date and a short random suffix: an ID used before, even in another chat, replays that earlier request or answers `REQUEST_ID_CONFLICT`. **On the board.** Inspect with `list_board_sessions` and `get_board_snapshot` and carry the same `sessionId` and a fresh `activeBoardId` as `expectedBoardId`. Native cards: `execute_board_action` (`listNativeModels`, `setModel`), `prepare_board_generation` with ordered board items or owned outputs, then one `generate_on_board` within the authorized budget. Board media tools: `list_media_tools` → `start_media_tool` → `get_media_tool_run`. Put headless results on a board with `add_media_to_board`. Keep mandatory safety settings and the live limits of each model. Prices, timings, endpoints and "free retry" claims in the sources are historical, not FlashBoards contracts. If a step this guide needs is not available or is refused (a tool error, `audio_not_configured`, a missing model or input), keep the finished assets, tell the person which step is blocked and why, and ask how to go on. A single generation that ends `failed` is not a blocked step: submit it once more with a new `generationAttemptId` (reworded if it may have hit the content checker), and stop to ask only if it fails again. A failed attempt is not always refunded, so the retry may be paid again and `maxCostUsd` caps each request on its own: count both against the budget the person approved. Never swap in another route on your own: no image model redrawing a layout or a frame, no `describe_media` in place of the frames it should check, no clip generated only to take its sound, no picture drawn where the guide takes a film frame. ## Source and related guides - [Original Learn guide](https://flashboards-h01.vercel.app/learn/formats-and-exhibition/ai-film-scene/skills/filmscene-sound.md) - Updated from the owner's skill sources of 2026-10-11. Read: `filmscene`, `filmscene-shots`, `reels-voice`, `reels-room-sound`. Source experiments and demo materials are historical; use the person's own approved media.