VibeSliding ✨

Transform ideas into stunning presentations in seconds. Write your content, pick a style, and let AI handle the rest — no design skills required.

Why VibeSliding? 🚀

• Save Hours — What used to take hours now takes minutes
• Professional Results — Every slide looks polished and cohesive
• Zero Learning Curve — If you can write markdown, you can create presentations
• Iterate Fast — Generate multiple variants instantly and pick your favorite

Features

• 🧪 DeepResearch — Turn a one-line idea into a researched, reviewable outline via Valyu
• 🎨 Two-Tone Style Transfer — Dark curtain plates (cover/transition/ending) + light teaching plates (content), with shared accents/motifs
• 🖼️ Slide References — Attach slide-specific photos or image globs in the outline
• 📚 Outline Articles — Declare source markdown/PDF references once at the top
• 🧠 Role-Aware Render — Routes each slide to the matching style plates by role (cover / transition / content / ending)
• 🔄 Multiple Variants — Generate several design options per slide
• 🎯 Selective Regeneration — Redo specific slides without starting over
• ⚡ Parallel Generation — Concurrent API calls with configurable concurrency
• 📄 Auto PDF — Combined slide PDF + speech-notes PDF
• 🎬 Narrated Video — TTS + ffmpeg turns slides and [Speech:] tags into MP4
• 🔌 Multi-Provider — OpenRouter (text + images) and Volcengine/Doubao Seedream (images)

Quick Start

pip install -r requirements.txt
cp .env.example .env   # add your API keys

Pipeline

Four tasks — research → outline → render → present:

# 1. Research: idea.md → research.md
python3 -m src.research.cli

# 2. Outline: research.md → outline_16.md, outline_25.md, outline_36.md
python3 -m src.outline.cli

# 3. Render: two-tone style refs → slide images + PDF
python3 -m src.render.style.cli   # base_noncontent → base_content → cover/transition/content
python3 -m src.render.cli

# 4. Present: slide PNGs + [Speech:] tags → narrated MP4 (requires ffmpeg)
python3 -m src.present.cli --output work/image_YYYYMMDD_HHMMSS

Then curate (delete variants you don't love) and polish (--page to regenerate individual slides).

💡 Steps 1–2 share ./work/ — edit files between steps. Style CLI writes five plates under style/ (style_base_noncontent.png, style_base_content.png, style_cover.png, style_transition.png, style_content.png). Render routes plates by slide role and writes images to work/image_YYYYMMDD_HHMMSS/ + PDFs to work/. Present writes presentation_video_YYYYMMDD_HHMMSS.mp4 to work/ (install ffmpeg first).

Usage Examples

# Render: pick a different outline, or skip interactive style prompts
python3 -m src.render.style.cli --outline work/outline_25.md
python3 -m src.render.style.cli --pick 1,1,2,1,3  # base_noncontent,base_content,cover,transition,content

# Render: multiple variants per slide
python3 -m src.render.cli --copy 4

# Render: longer outline + selective regeneration
python3 -m src.render.cli --outline work/outline_36.md --page "1,3,5-7"

# Render: rebuild PDF after curating (no API calls)
python3 -m src.render.cli --pdf-only --output work/image_20260520_220006
python3 -m src.render.cli --pdf-only --output work/image_20260520_220006 --variant 1

# Render: explicit style plates + article references
python3 -m src.render.cli --style "style/*.png" --article "docs/*.pdf"

# Present: TTS + video from curated slides (MiniMax TTS)
python3 -m src.present.cli --output work/image_20260520_220006 --variant 1

# Present: clone your voice from a short reference recording (clean WAV/MP3)
python3 -m src.present.cli --output work/image_20260520_220006 \
  --reference-audio work/voice_sample.wav

# Present: remux only after TTS MP3s already exist (no API calls)
python3 -m src.present.cli --output work/image_20260520_220006 --mux-only

CLI Reference

src.research.cli

Option	Description
--work	Work directory (default: work/)
--mode	Valyu mode: fast / standard / heavy / max
--categories	Comma-separated datasource categories (see below)
--valyu-api-key	Valyu API key override
--resume	Resume polling from research_state.json
--task-id	Resume a specific task ID
--fresh	Start new task even if state file exists

Categories: research, healthcare, patents, markets, company, economic, predictions, legal, politics, cybersecurity, transportation

src.outline.cli

Option	Description
--work	Work directory (default: work/)
--slides	Content slide counts (default: 16,25,36)
--api-key	OpenRouter API key override
--txt-model	Text model override
--proxy	HTTP/HTTPS proxy for OpenRouter

src.render.style.cli

Option	Description
--work	Work directory (default: work/)
--outline	Outline file (default: work/outline_16.md)
--candidates	Candidates per stage (default: 4, max: 12)
--pick	Pre-select indices: base_noncontent,base_content,cover,transition,content
--provider	openrouter or volcengine
--api-key	API key override
--proxy	HTTP/HTTPS proxy (OpenRouter only)

src.render.cli

Option	Description
--work	Work directory (default: work/)
--outline	Outline file (default: work/outline_16.md)
--style	Style image(s)/glob, repeatable (default: style/*.png)
--copy	Variants per slide (default: 1)
--page	Pages to generate, e.g. 1,3,5-7
--article	Article path(s)/glob, repeatable (.pdf, .md)
--output	Output dir (default: work/image_YYYYMMDD_HHMMSS/); PDFs go to --work
--provider	openrouter or volcengine
--api-key	API key override
--proxy	HTTP/HTTPS proxy (OpenRouter only)
--balance-only	Print OpenRouter credits and exit
--no-balance	Skip credits line after run
--pdf-only	Rebuild PDFs from existing PNGs in --output dir (no API calls); speech PDF always uses first variant per slide
--variant	With --pdf-only: variant filter for slides PDF only, e.g. 1 or 1,2

Credits: After each OpenRouter run, remaining credits are printed unless --no-balance. Requires an OpenRouter Management API key — set OPENROUTER_MANAGEMENT_API_KEY or [openrouter] management_api_key in .env.

src.present.cli

Option	Description
--work	Work directory (default: work/)
--outline	Outline file (default: work/outline_16.md; falls back to snapshot in --output)
--output	Render image directory with slide_p##_v##.png (required)
--page	Slides to include, e.g. 1,3,5-7
--variant	Variant filter, e.g. 1 or 1,2
--api-key	MiniMax API key override
--tts-model	MiniMax TTS model (default: speech-2.8-hd)
--voice	TTS voice (default: Chinese (Mandarin)_Lyrical_Voice)
--reference-audio	WAV/MP3/FLAC of your voice for MiniMax voice cloning
--voice-id	Saved MiniMax voice ID (alternative to --reference-audio)
--mux-only	Skip TTS; mux existing slide_p##_v##.mp3 files (no API calls)
--silent-seconds	Duration for slides without speech (default: 3.0)

Prerequisite: ffmpeg and ffprobe must be on your PATH.

Voice clone: --reference-audio uploads reference audio and clones voice dynamically via MiniMax.

Input/Output

Work Directory

style/                                         # style-reference plates (default --style glob)
├── style_base_noncontent.png                  # dark curtain base (cover/transition/ending)
├── style_base_content.png                     # light content base (teaching slides)
├── style_cover.png                            # cover-slide reference (dark)
├── style_transition.png                       # transition/roadmap reference (dark)
├── style_content.png                          # content-slide reference (light)
└── style_candidates/                          # all candidates + contact sheets
work/
├── idea.md                                    # seed: title, audience, core message
├── research.md                                # DeepResearch report with citations
├── research_state.json                        # (transient) in-progress task state
├── source.md                                  # optional custom source material
├── style_base.md                              # shared style + narrative scaffold
├── outline_16.md                              # 16 content-slide outline
├── outline_25.md                              # 25 content-slide outline
├── outline_36.md                              # 36 content-slide outline
├── image_YYYYMMDD_HHMMSS/                     # render output (slide PNGs)
├── presentation_slides_YYYYMMDD_HHMMSS.pdf    # combined slide deck
├── presentation_speech_YYYYMMDD_HHMMSS.pdf    # A4: first image per slide + speech notes
└── presentation_video_YYYYMMDD_HHMMSS.mp4       # narrated slide video

Outline Format

# PPT Outline: My Presentation

[Articles: @docs/research.md, @docs/report.pdf]

---

## Slide 1: Introduction
- **Key point:** Why this matters
- Core insight: One takeaway the audience should remember
[Reference: photos/founder.png]
[Visual: Split-screen hero; light content background; Reference style: style_content.png, style_base_content.png]
[Speech: Conversational presenter narration for this slide]

---

## Appendix: Global Visual Requirements
- **Theme:** Two-tone deck — white/light content slides; dark curtain for cover, transitions, ending. Accents: Primary #2563EB, Accent #10B981
- **Fonts:** Title / body families and sizes

Tag	Purpose
[Visual: ...]	Layout, composition, diagrams, icons, motifs; name the correct style plates for the slide role
[Speech: ...]	Presenter narration (used by present + speech PDF)
[Reference: ...]	Slide-specific image refs (place immediately before [Visual:]; leading @ accepted)
[Articles: ...]	Top-of-outline text reference declarations
## Appendix: ...	Deck-wide text constraints only (theme/hex, fonts/sizes; state two-tone backgrounds)

Include 3–6 transition slides (titles prefixed Roadmap:) with progress bar N/total markers. Render attaches dark plates to cover/transition/ending and light plates to content slides.

Render Output

work/
├── image_YYYYMMDD_HHMMSS/
│   ├── outline_16.md                    # outline snapshot used for this run
│   ├── slide_p01_v01.png              # Slide 1, variant 1
│   ├── slide_p01_v01.mp3              # TTS audio (after present)
│   ├── slide_p01_v02.png              # Slide 1, variant 2
│   └── slide_p02_v01.png              # Slide 2, variant 1
├── presentation_slides_YYYYMMDD_HHMMSS.pdf   # all slides in one PDF
├── presentation_speech_YYYYMMDD_HHMMSS.pdf   # A4: first image per slide + speech notes
└── presentation_video_YYYYMMDD_HHMMSS.mp4    # narrated MP4

Configuration

Copy .env.example → .env and fill in your keys. INI-style sections:

Section	Used by	Key settings
(preamble)	All	provider, max_concurrent, proxy
[openrouter]	Outline / Render (when provider = openrouter)	api_key, management_api_key, img_model, txt_model, use_proxy
[volcengine]	Render (when provider = volcengine)	api_key, img_model, txt_model, use_proxy
[minimax]	Present (+ future text/image)	api_key, img_model, txt_model, use_proxy, optional tts_model, tts_voice
[valyu]	Research	api_key, mode, categories, use_proxy

• Research → [valyu] for DeepResearch
• Outline → txt_model from the active provider section (falls back across [openrouter], [volcengine], [minimax])
• Render → img_model from the active provider (src.render.style.cli + src.render.cli)
• Present → MiniMax TTS via [minimax] tts_model / tts_voice or --tts-model / --voice (defaults: speech-2.8-hd, Chinese (Mandarin)_Lyrical_Voice) with optional voice cloning
• Global proxy applies when a section's use_proxy = true; section-level proxy overrides are still supported
• ark_api_key in the preamble aliases [volcengine] api_key

License

See LICENSE file for details.

---

Happy Sliding! 🎨✨ Transform your markdown into beautiful presentations in minutes, not hours.