# Speak in Progress — Complete System & Methodology Specification

> Official LLM & AI agent documentation for the Speak in Progress spoken fluency production suite.

## 1. Overview
Speak in Progress is a dedicated desktop YouTube recording studio and deliberate practice framework engineered specifically for English learners documenting their communicative fluency journey.

Rather than relying on passive listening, traditional vocabulary apps, or grammar drills, Speak in Progress couples continuous teleprompter reading with instant 90-second unscripted memory retellings, tracking daily stumble rates and script independence over time.

## 2. The 5-Step Screen-Recording Workflow
Each daily practice session follows a standardized sequence structured for OBS Studio and Loom screen-recording:

### Step 1: Voiceover Intro (15 seconds)
A high-energy opening hook introducing the topic, day number, and personal progress goal. Designed to establish vocal projection and engage viewers within the first 10 seconds of a YouTube video.

### Step 2: Story Reading (2–3 minutes)
A continuous 5-paragraph leveled narrative read aloud from the teleprompter:
- Conversational Cadence: 130 WPM (28 px/s drift rate)
- Presentation Cadence: 145 WPM (34 px/s drift rate)
- Executive Cadence: 160 WPM (42 px/s drift rate)
Features an eye-line focus beam, paragraph focus mode, and fullscreen recording view (`F`).

### Step 3: Key Expressions (1–2 minutes)
Deep dive into 3 high-impact idioms or collocations contextualized within the daily story:
- Expression 1: Core conceptual collocation
- Expression 2: Action or momentum phrase
- Expression 3: Reflective or strategic verbal idiom
Includes contextual definitions, usage examples, and vocal intonation guidance.

### Step 4: Unscripted Retelling (90 seconds)
The script scrolls out of view. The speaker must verbalize the story from memory within a 90-second countdown using 4 structured cognitive anchor triggers. This forces the brain to construct grammar dynamically under gentle time pressure, bridging the gap between passive comprehension and spontaneous speech.

### Step 5: Private Log (1 minute)
Personal neuro-linguistic self-assessment:
- Script Independence Score (1 to 5 stars)
- Stumble & Friction Notes (phonemes, hesitation points)
- Win of the Day (pronunciation breakthrough, phrasing milestone)
Stored in private Cloud Firestore.

## 3. Tool Suite
- **Interactive Teleprompter Studio**: Real-time adjustable speed, audio practice recorder, fullscreen screen-recording canvas.
- **16:9 Thumbnail Studio**: 2K (2560×1440) and FHD (1920×1080) YouTube thumbnail builder with bold editorial typography, customizable highlight words, and instant PNG download.
- **Video Intro & Channel Banner Suite**: 3–4 second animated video stings with real audio chords, 2560×1440 channel banners with desktop/mobile TV safe zones, and SVG branding logos.
- **Voice & Tone Guidelines**: Creator vocal pacing, articulation guidelines, and production standards.

## 4. Privacy & Authorization Model
- Public Landing Page: Fully indexed by search engines (`index, follow`).
- Internal Studio: Strictly restricted to authorized creator credentials.
- Once authenticated into the studio, the runtime automatically toggles search robots to `noindex, nofollow, noarchive, nosnippet`, ensuring personal recordings and draft logs are never indexed.
- Other accounts attempting to log in receive a polite notice directing inquiries to `support@mdselim.com`.

## 5. Contact & Support
- Founder & Creator: Md. Selim
- Support & General Inquiries: support@mdselim.com