Ninja Listening Trainer

Ninja Listening Trainer is a specialized browser extension designed to bridge the gap between textbook pronunciation and natural, real-world spoken language. Whether you are practicing shadowing, analyzing fast native speech, or conducting acoustic perception research, this tool gives you precise control over media playback and step-by-step training sequences.

• Instant Marking: Set precise Start and End points on any YouTube video or direct local audio file with millisecond accuracy.
• Custom Step Sequences: Seamlessly iterate through dynamic playback speed programs—from slow anchor speeds ($0.6\times$) up to time-compressed challenge speeds ($2.0\times$).
• Automatic Subtitle Sync: Toggle captions on or off automatically per step to force acoustic listening before relying on textual support.

• Optional VAS (Visual Analogue Scale) Ratings: Turn on Research Mode to assess your acoustic and phonological perception clarity at each step.
• Phonetic-Level Feedback: Measure how clearly you perceive individual sounds, phonemes, and syllables—not just general word boundaries.
• One-Click Data Export: Save your session ratings, timestamps, and playback configs into clean CSV or JSON files for statistical analysis (ANOVA, Pandas, R) or study logging.

• Built-in Experimental Protocols: Instant 1-click access to the Ninja Protocol ($0.6\times$ Anchor $\rightarrow 2.0\times$), Pure Compression, and Static Control modes.
• Load Custom Configs: Import/export standardized lesson files (.json) to share study protocols with students, study groups, or experiment participants.
• Local Media Support: Works natively on YouTube (youtube.com/watch) as well as local audio and video files opened directly in Firefox.

1. Mark Your Target: Navigate to a video/audio segment and click Mark start and Mark end.
2. Choose Your Preset: Select the Ninja Protocol or load a custom step configuration.
3. Run Sequence: Click ▶ Run sequence to begin your step-by-step listening practice.
4. (Optional) Check Collect VAS Ratings in the controls panel if you want to record perceptual clarity scores after each iteration and export your dataset.