# Technical information

This page records the technology used for the 57-second main story at the beginning of the video. It does not describe the entire landscape-format HOWTO program. Model and service names reflect the production record at the time.

## Final main video

- 57 seconds
- 19 shots
- 1280 × 720
- 24 fps
- Audio included

## Planning and storyboard

- Planning: Codex
- Excel storyboard: Python, openpyxl, and Pillow
- Basic unit: one three-second shot
- Storyboard checks: fixed character positions, speaker, displayed paper, month differences, and Tessy’s final expression and close-up

## Video generation

- Google Gemini API
- Model: `gemini-omni-flash-preview`
- All 19 shots were generated
- Audio contained in generated clips was removed; dialogue and music were assembled separately

The model ID used to create the character and reference images was not recorded, so it is not guessed or listed here.

## Dialogue voice

- Google Gemini TTS
- Model: `gemini-3.1-flash-tts-preview`
- Pride: Orus
- Kakyo: Algenib
- Tessy: Achird

## Music

- Distributor: [DOVA-SYNDROME](https://dova-s.jp/)
- Track: [奇妙な案内人 (Kimyouna Annainin)](https://dova-s.jp/bgm/detail/12153)
- Composer: MATSU
- Terms: [DOVA-SYNDROME License](https://dova-s.jp/help/articles/license/)
- Acquisition record date: July 29, 2026
- Placement: from 54.4 seconds to the end, approximately 2.6 seconds

To thank the services and creators used in the production, the distributor, track, composer, and license-information link are shown here.

## Editing and audio

- Python
- FFmpeg
- Removal of generated-clip audio
- Synchronization of dialogue and non-tonal room ambience with video
- Mixing of the ending music
- Dialogue ends at 53.5 seconds, the image cuts to Tessy at 54.0 seconds, and the music begins at 54.4 seconds

## Quality checks

- Full video decoding
- Full audio decoding
- Internal-frame checks across all 19 shots
- Match between dialogue and visible speaker
- Character position and eyeline checks
- Tessy’s final expression, camera move, and music start timing

## Major production corrections

- Corrected the mistaken direct client-to-engineer exchange to a PM, PMO, and engineer structure.
- Fixed the three people at left, center, and right in the same command room so they did not appear back-to-back.
- Replaced the initial “OH NO” concept with royalty-free music that matched the intended emotion.
- Measured the completed dialogue to prevent the music from entering too early.
- Changed Tessy so he was already confused at the start of the close-up instead of becoming surprised midway through it.