Skip to content

Scene Modes 📤

Switch between multiple creation modes at any time, supporting various scenarios such as video dubbing, translation, and narration to meet the needs of different users.

Configuration Options

Video

  • Local Path: C:\Users\home\Desktop\test.mp4

  • Online Link: Youtube, Tiktok

TIP

The file will be automatically downloaded to the webapp\download directory on the first run.

Subtitles

  • Local Path: C:\Users\home\Desktop\test.srt

TIP

Optional. If provided, subtitle recognition will prioritize this file. This is especially useful for long texts, significantly reducing recognition time.

Language

10 languages are built-in by default.

locale.json
json
[
  {"Locale": "zh-CN", "Name": "中国", "Country": "China"},
  {"Locale": "en-US", "Name": "美国", "Country": "United States"},
  {"Locale": "en-GB", "Name": "英国", "Country": "United Kingdom"},
  {"Locale": "ja-JP", "Name": "日本", "Country": "Japan"},
  {"Locale": "ko-KR", "Name": "韩国", "Country": "South Korea"},
  {"Locale": "de-DE", "Name": "德国", "Country": "Germany"},
  {"Locale": "fr-FR", "Name": "法国", "Country": "France"},
  {"Locale": "pt-BR", "Name": "巴西", "Country": "Brazil"},
  {"Locale": "es-ES", "Name": "西班牙", "Country": "Spain"},
  {"Locale": "ru-RU", "Name": "俄罗斯", "Country": "Russia"},
]

TIP

Switching regions will trigger translation of the preview text.

Modes

Dubbing Mode

Dub the video according to the specified alignment method, adapting to different usage scenarios. Below are three strategies:

  • Video: Dub according to the length of the video, with accelerated dubbing.

    Video durations are often limited, requiring key information to be conveyed in a short time. Alignment typically accelerates dubbing to fit within the video duration, allowing a complete product feature or brand story to be told.

  • Audio: Dub according to the length of the audio, with the remaining duration frozen on the last frame until the end.

    When the video length is insufficient, and the dubbing length far exceeds the audio length, this mode is mainly used with Stable Diffusion for post-processing image filling.

  • Adaptive: Adjust the speed of both video and dubbing within a reasonable range according to a certain threshold.

    Suitable for most scenarios.

Translation Mode

Translate the video language into another language, strictly aligning voiceprint <-> subtitles.

Usage Limitations: The following scenarios may require manual intervention:

    1. When there are multiple speakers and overlapping speech occurs.
    1. Due to differences in speech speed between languages, when there are significant variations in speech tempo.

Reconstruction

Example: When a user uploads test.mp4,

All related artifacts of test.mp4 will be deleted, equivalent to starting over. Please proceed with caution.

Directory Files

    1. All files generated during the process will be saved in the webapp\temp\test directory.
    1. For terminology explanations, see the "Directory Files" section.

Markers

Example: When a user uploads test.mp4 and marks it as 001,

Previously generated videos and drafts will not be overwritten.

Output Artifacts

    1. Draft Directory: {filename}_{marker}
    1. File Format: {filename}_{marker}.mp4