-
Notifications
You must be signed in to change notification settings - Fork 0
Feature Reference
Anubha Parashar edited this page Jun 15, 2026
·
1 revision
| Feature | Description |
|---|---|
| Resume-aware questioning | Generates questions based on candidate resume content |
| JD-aware interview mode | Aligns questions with target job description |
| Coding round | Supports technical problem solving and evaluation |
| Face-to-face round | Simulates interviewer-style probing questions |
| Candidate report | Produces structured evaluation output |
| Feature | Description |
|---|---|
| Adaptive training catalog | Creates targeted training topics from weaknesses |
| Concept slides | Shows definition, relevance, interviewer rationale, checkpoint |
| Live browser teaching | Fast preview with installed browser voice |
| XTTS generated teaching | Uses authorized voice sample to generate lecture audio |
| Digital-human video | Uses SadTalker/Wav2Lip to render the presenter |
| Feature | Description |
|---|---|
| Authorized presenter photo | Uses a selected/consented presenter source image |
| Authorized voice sample | Uses 30–90 second user-approved voice sample |
| Background generation | Runs audio and video generation without blocking UI |
| Cached prefetch | Reuses previous generated outputs |
| YouTube-style player | Play/pause, seek, speed, volume, forward/back controls |
| Final MP4 download | Downloads video + generated audio together |
| Feature | Description |
|---|---|
| Consent checkbox | Requires explicit model terms and voice-use acknowledgement |
| No silent fake fallback | Does not silently switch to automated voice |
| User-approved fallback | Automated voice can be used only after user approval |
| Separate evidence capture | Candidate capture is separate from digital presenter generation |
ASHU Mentor AI Studio · Adaptive Smart Human-like Unit Mentor AI Studio
Home · Demo & Live App · Architecture · Features · Security & Consent
Maintained by @dranubhaparashar