is there way to add a emotional intelligence like detect the vocal prosody

Thanks for reaching out, Sudharshan. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.

i am creating multilingual voice agent. livekit packagees or plugins i should defenenately add to the agent like selirio

Thanks — I’m glad you clarified. You meant “silero” (the VAD plugin). I’ll highlight the must-have LiveKit packages/plugins for a multilingual voice agent and ask a couple quick questions so I can tailor recommendations. Short answer — essentials to include: • Core Agents SDK ◦

livekit-agents

(foundation for agent logic and plugin loading) • VAD (Voice Activity Detection) ◦

livekit-plugins-silero

— highly recommended for accurate start/stop speaking detection and natural turn-taking • Multilingual STT (Speech-to-Text) ◦ A plugin for a multilingual STT provider (example:

livekit-plugins-deepgram

) that supports automatic language detection or

language="multi"

• TTS (Text-to-Speech) ◦ High-quality multi‑language TTS (examples:

livekit-plugins-elevenlabs

or

livekit-plugins-rime

) so your agent can speak naturally in each language • LLM integration ◦

livekit-plugins-openai

(or your preferred LLM plugin) for the conversational brain • Turn detection ◦

livekit-plugins-turn-detector

with the

MultilingualModel

for better turn-taking across languages Useful optional additions: • Noise cancellation / audio enhancement plugin to improve STT/TTS quality in noisy environments • Sentence-boundary or text-processing helpers (for long TTS outputs or language-specific punctuation handling) • Any language-specific tokenizers or normalization tools if you plan to support languages with special segmentation needs (e.g., Chinese, Japanese) Next, a couple of quick questions so I can give precise setup steps: 1. Which STT and TTS providers do you plan to use (or want to evaluate)? (e.g., Deepgram, Whisper, Google, ElevenLabs, Rime) 2. Are you doing automatic language detection/switching, or will the language be known up-front per session/user? 3. Do you want prosody/emotion detection as well (we can recommend approaches/plugins for that)? If you tell me those three things I’ll give: • Exact package install commands and an example AgentSession config • Recommended plugin settings for multilingual STT/TTS and turn detection • If desired, a short plan to add prosody/emotion detection (third‑party or custom feature capture) Sources: Building the all-in-one platform for voice AI agents | LiveKit turn detector plugin | LiveKit Documentation | Build a multilingual voice agent that automatically switches languages | Li... | Agents Framework | livekit/livekit_composite | DeepWiki