screencastgen

Pipeline Types

Request and result dataclasses for pipeline invocation.

Source: screencastgen/pipelines/types.py

Request Hierarchy

BasePipelineRequest
    └── TTSRequest
        ├── AudioPipelineRequest
        ├── HighlightPipelineRequest
        └── LipsyncPipelineRequest

VisualizationPipelineRequest
RenderedVisualClip

`BasePipelineRequest`

Field	Type	Default	Description
`pdf`	`str`	—	Input document path
`output`	`str`	`None`	Output filename
`output_dir`	`str`	`"audio"`	Output directory
`language`	`str`	`"en-US"`	Language code
`status_file`	`str`	`None`	Tracker JSON path
`clean`	`bool`	`False`	Clear previous state
`verbose`	`bool`	`False`	Verbose output

`TTSRequest`

Field	Type	Default	Description
`backend`	`str`	`"qwen"`	TTS backend name
`device`	`str`	`"auto"`	Compute device
`voice`	`str`	`None`	Voice identifier
`model`	`str`	`None`	Model name or size
`ref_audio`	`str`	`None`	Reference audio for cloning
`ref_text`	`str`	`None`	Reference transcript
`tts_server_url`	`str`	`None`	Remote TTS server URL
`aligner`	`str`	`"whisperx"`	Alignment provider
`tts_concurrency`	`int`	`1`	Number of chunks to synthesize in parallel

`AudioPipelineRequest`

Field	Type	Default	Description
`no_concat`	`bool`	`False`	Skip concatenation step

`HighlightPipelineRequest`

Field	Type	Default	Description
`format`	`str`	`"epub"`	Output format: `epub` or `mp4`
`font_size`	`int`	`32`	Text font size
`resolution`	`str`	`"1280x720"`	Video resolution
`fps`	`int`	`24`	Frames per second

`LipsyncPipelineRequest`

Field	Type	Default	Description
`format`	`str`	`"reader"`	Output format: `reader`, `mp4`, or `epub`
`ref_video`	`str`	—	Reference face video
`lipsync_provider`	`str`	`"auto"`	Lip-sync provider
`face_position`	`str`	`"bottom-right"`	Presenter position
`face_scale`	`float`	`0.22`	Presenter scale for docked corner layouts
`latentsync_preset`	`str`	`"quality"`	Quality preset

reader builds hosted assets and a standalone offline ZIP. The epub option is a text-and-narration accessibility export without presenter video.

`VisualizationPipelineRequest`

Field	Type	Default	Description
`prompt`	`str`	—	Concept prompt to visualize
`output`	`str \\| None`	`None`	MP4 filename, default `visualization.mp4`
`output_dir`	`str`	`"audio"`	Directory for source, metadata, and render output
`provider`	`str`	`"manimgl"`	Renderer provider: `manimgl` or `manimce`
`duration_seconds`	`int`	`30`	Target animation duration
`resolution`	`str`	`"1280x720"`	Render resolution
`fps`	`int`	`24`	Frames per second
`style`	`str`	`"clean"`	`clean`, `chalkboard`, `blueprint`, or `minimal`
`audience_level`	`str`	`"general"`	Labeling hint for generated scene
`iteration_of_job_id`	`str \\| None`	`None`	Optional reference to a prior visualization job
`timeout_seconds`	`int`	`300`	Renderer subprocess timeout
`max_output_bytes`	`int`	`512 MiB`	Output size guardrail

`RenderedVisualClip`

Metadata describing a rendered visualization clip: path, duration, fps, dimensions, source prompt, and generated source code.

Result

`PipelineRunResult`

Field	Type	Description
`exit_code`	`int`	`0` means success
`output_name`	`str`	Output filename
`output_path`	`str`	Full output path
`error_message`	`str`	Error description when failed
`metadata`	`dict`	Pipeline-specific metadata stored by web jobs

Utility Function

`coerce_request(request_type, value)`

Converts an argparse.Namespace or dict into the requested dataclass type. Used by both CLI and Pipeline Tasks.

Dependencies

Pipeline Types
├── dataclasses (stdlib)
└──▶ consumed by Audio Pipeline
     ├──▶ Highlight Pipeline
     ├──▶ Lipsync Pipeline
     ├──▶ Visualization Pipeline
     ├──▶ CLI
     └──▶ Pipeline Tasks

screencastgen

Pipeline Types

Request Hierarchy

BasePipelineRequest

TTSRequest

AudioPipelineRequest

HighlightPipelineRequest

LipsyncPipelineRequest

VisualizationPipelineRequest

RenderedVisualClip