A voice-enabled AI system called Conversation Coach has been developed to help managers practice difficult workplace conversations, addressing challenges in low-latency interaction, adaptive bot personalities, and personalized feedback. The system offers an end-to-end speech-to-speech model that is faster and cheaper but less capable in reasoning, and a cascaded approach that uses separate speech recognition, LLM, and text-to-speech components, which provides better coaching quality. Separately, a real-time voice AI receptionist named Maya was built for a clinic, handling tasks like checking doctor availability, booking appointments, and providing medical triage, utilizing a stack that includes LiveKit, Groq Whisper, Groq LLM, and Cartesia TTS for low-latency interactions. AI
IMPACT These projects highlight advancements in real-time voice AI for specialized applications, potentially improving user interaction in professional training and customer service scenarios.
RANK_REASON The cluster contains two distinct AI system development projects, one described in an arXiv paper and the other as a personal project blog post.
- alphaXiv
- arXiv
- CatalyzeX
- Connected Papers
- Conversation Coach
- CORE Recommender
- DagsHub
- Gotit.pub
- Hugging Face
- Litmaps
- ScienceCast
- scite Smart Citations
- Bengaluru
- Cartesia TTS
- Groq
- Groq LLM
- Groq Whisper
- LiveKit Cloud
- Silero VAD
- SwasthyaCare Clinic
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →