DHXpresso

We warmly invite you to the DHXpresso Meeting, our monthly meeting for the Digital Humanities community. Our next edition on Friday, the 16th of January at 11 o’clock will feature a presentation on the topic “Enhancing Accessibility Through Audio-Visual Transcription and Translation Service (Voice-AI)”.

In this session, we’ll dive into: This talk examines the deployment of Voice AI using the Whisper model, tracing its transition from theoretical research to practical, wide-scale application. The session demonstrates cURL-based integration, discusses fine-tuning strategies, and showcases the ‘Voice Live AI’ beta for real-time transcription. Furthermore, the presentation analyzes persistent challenges such as speaker diarization and latency, ultimately offering a strategic roadmap for optimizing service performance and stability.

Our speaker is Narges Lux from GWDG.