Trainexus
VideoAnalytics
Trainexus
VideoAnalytics
Upload a video and receive a factual evaluation within minutes: body language, eye contact, voice, speaking rate, filler words — with concrete pointers on what to work on.
Processed in the EU · Raw video deleted after analysis · Results are yours alone
From your recording we produce an annotated video plus charts so you can trace eye contact, body posture, speech rate and more.
VideoAnalytics shows you, based on your own recording, how your talk comes across: eye contact, gestures, posture, voice and your relation to the presentation are made visible – and translated into clear feedback.
A tool by Trainexus for anyone who wants feedback on conversations, dialogue and presentation situations — from preparing a talk to practising difficult conversations.
Record your webcam and optionally your screen, then upload for analysis.
No software installation required. The compatibility check verifies in advance whether camera, microphone and browser audio are working.
Start recordingSubmit an existing recording (MP4, WebM, MOV) for analysis.
Upload an existing video file. The analysis starts automatically and you will receive your feedback by email.
Upload videoRecord your presentation via webcam or upload an existing video. Optionally, you can also record your screen at the same time.
The system detects eye contact, gestures, posture, pauses, volume and speaking rate. Once the analysis is ready, you receive the result link by email – you can close the page in the meantime.
You will receive an email with a link to your results including interactive charts, a transcribed text version and optionally an AI coaching report.
Any situation where you speak to a camera or an audience — repeatable as often as you like, with nobody watching.
Record the same talk several times and compare how eye contact, pace and filler words develop.
Check whether your core message lands clearly and in a well-structured way in a short time.
Review your own recording for clarity and camera contact before sending it.
Answer typical questions as private practice and refine how you speak.
Practise a short, clear introduction that does not feel overloaded.
Spot rushed delivery, monotone passages and long pauses before the video goes live.
Analyse voice, pace, pauses and filler words — plus gaze and body language for video.
Practise transitions, announcements and handling longer pauses.
Explain complex content concisely, in a structured and comprehensible way.
Review an update or address before it goes out to your team.
Communication training barely scales with in-person trainers. Video analysis gives every individual immediate feedback — trainer time then goes where it is genuinely needed.
Inconsistent conversation quality costs deals and ties up many hours of coaches and team leads.
One-to-one communication training is expensive and hard to scale across a large workforce.
Complex, regulated product messages must be delivered reliably — in-person training costs travel and trainer time.
Presentation and pitch quality directly affects win rates and client satisfaction.
Pace, pauses and filler words make conversations harder and increase handling time.
Participants keep practising between supervised sessions with standardised feedback.
Videos that are hard to follow have to be reshot — costing coordination and working time.
For voluntary training of interviewers. Assessing candidates with it is explicitly not permitted.
Access works through a personal link. Whoever holds it can upload — no account, no sign-up.
Tell us briefly what you would like to use the analysis for. We will send you a link for one free trial analysis.
Organisations book a quota, for example 7 links with 4 analyses each. You distribute the links yourself — to whom is entirely your decision.
Your organisation only sees how many analyses were used in total — never who uploaded which video. Results go to the individual alone.
After the analysis you receive an interactive feedback dashboard. Here are examples of the different analysis sections:
Skeleton overlay shows posture, gestures and gaze direction in real time
Interactive timeline: jump straight to moments with lots of gesturing, little eye contact or a change in voice.
Pointers on the clarity, structure and impact of your talk.
Colour-coded improvement suggestions directly in the transcript
Eye contact, speaking rate, gestures and pauses – clearly summarised in collapsible cards.
Clear, plain-language pointers: what already works well and what you can concretely work on.
A short calibration so your eye contact is detected more reliably.
Set start and end time so only the relevant section is analysed — no distorted data from walking in
Per analysis section: thumbs up/down + comments. Your feedback helps us improve the software
The analysis describes only observable behaviour and measured values – for example camera contact as a percentage of speaking time, speaking rate or pitch variation. It does not attribute feelings, moods or inner states to you; this also applies to the AI-generated texts.
Facial-expression, emotion and nervousness/confidence analysis: this data is not collected or analysed – because of Article 5(1)(f) of the EU AI Act (Regulation (EU) 2024/1689), which prohibits inferring emotions in education and workplace settings; the prohibition has applied since 2 February 2025.
The use of this service is entirely voluntary. There is no obligation to record or upload videos for analysis. Optional features (e.g. AI coaching) must be actively enabled.
All uploaded videos and analysis results are automatically deleted after no more than 14 days. After that, neither the video nor the analysis data remains accessible. You can also trigger immediate deletion yourself at any time – via the “Delete data” button on your personal feedback page. All files are removed instantly; any copy on the analysis server is deleted within about 15 minutes.
Video analysis runs on dedicated servers within the EU. Uploaded videos never leave the EU; the raw video is deleted once the evaluation is complete.
This website uses no tracking, no advertising or analytics cookies and no external analytics services. All styling and script libraries are served directly from our own server – no content is loaded from external servers (e.g. CDNs), so your IP address is not exposed to third parties either. The only cookies used are a strictly necessary one for your language choice and a session cookie that protects forms. Beyond that, no personal data is collected apart from the email address required to deliver your feedback.
The analysis combines video, speech and voice features. Results serve self-reflection and are not an assessment of you or your performance. Technical overview:
| Category | Technology | Purpose |
|---|---|---|
| Body Pose & Gesture | MediaPipe (Google) | Body pose (skeleton landmarks), face mesh (468 points), iris tracking |
| Speech Recognition | faster-whisper (Systran) | Speech-to-text (transcript), word-based filler detection |
| Gaze Direction | L2CS-Net (ResNet-34) | Gaze angle estimation (yaw = horizontal rotation, pitch = vertical tilt) from the face crop |
| Voice Analysis | openSMILE (audEERING) | Acoustic features following the eGeMAPS standard (extended Geneva Minimalistic Acoustic Parameter Set): pitch, loudness, jitter = pitch fluctuation, shimmer = loudness fluctuation, HNR = Harmonics-to-Noise Ratio (ratio of harmonic content to noise) |
| Voice Quality | Parselmouth/Praat | Voice quality metrics based on Praat (the de-facto standard tool of phonetics research) |
| Filler Words (audio-based) | Eigenes Verfahren (Trainexus, auf eGeMAPS-Basis) | Custom method: detects “uh” / “um” from acoustic features (eGeMAPS) instead of from the transcript — more reliable than text-only detection |
| Emphasis & Three-Channel Coherence | Eigenes Verfahren (Trainexus) | Custom method: measures whether vocal emphasis, gesture and pause align while speaking (three-channel coherence) |
| Orientation Toward Presentation | Eigenes Verfahren (Trainexus) | Custom method: infers from hand, gaze and body direction when the speaker is facing the audience vs. the presentation (slides/board) |
| AI Coaching | Claude (Anthropic) | Speech-quality analysis and personalised coaching based on transmitted text and metrics (no video/audio) – contractually under a Zero-Data-Retention agreement: content is not stored and not used for AI training. |
| Video Processing | OpenCV + FFmpeg | Per-frame processing, video encoding |
| Charts | Chart.js | Interactive timeline visualisation |
| GPU Acceleration (graphics-card computing) | NVIDIA CUDA 12.1 | AI-inference acceleration on NVIDIA graphics cards |
| Framework | FastAPI + Python | Backend interface (API = Application Programming Interface) |
Trainexus UG (haftungsbeschränkt)
Zum Wachtfels 17
91241 Kirchensittenbach
Paul Dölle
Trainexus UG (haftungsbeschränkt)
Zum Wachtfels 17, 91241 Kirchensittenbach, Germany
Trainexus UG (haftungsbeschränkt), represented by Paul Dölle (managing partner). Register court: Amtsgericht Bayreuth, HRB 8503.
Bavarian State Office for Data Protection Supervision (BayLDA), Promenade 18, 91522 Ansbach, Germany