Speech recognition based on OpenAI Whisper for hands-free operation. You dictate your query and the AI responds directly.
Click to speak
Transparency notice: this demo is a simulation using prepared sample data and does not show output from a live AI model. Labelled in line with Art. 50 of Regulation (EU) 2024/1689.
Speech is converted to text while you are still speaking.
Automatic language detection and transcription in over 50 languages including English and German.
Whisper runs locally on your GPU, which makes transcription faster than on the CPU.
All audio data stays on your server. No cloud services, no external APIs.
Audio filters improve recognition in noisy environments.
The AI understands the context of previous conversations for more natural dialogue.
Automatic transcription of meetings and conferences with speaker identification.
Dictate emails, reports, and notes directly into your applications.
Users with limited mobility can operate the AI by voice.
The demo shows how Whisper transcribes spoken queries locally and passes them to the AI.
