Emoaid:Chatbot for Sensitive Souls

Speak Your Heart — We Listen, Understand, and Heal

Comment

Inspiration

Mental health often goes unspoken, especially for those who are highly sensitive, introverted, or struggling in silence. We wanted to build something soft, safe, and supportive — a voice-enabled chatbot that listens without judgment and responds with compassion. This project is our way of making tech more emotionally intelligent and accessible.

What it does

EmotionAid is a multilingual, emotionally intelligent chatbot designed for sensitive souls. It can: Record and transcribe your voice Understand your emotional state using mood tagging Respond kindly based on detected feelings (e.g., loneliness, stress, heartbreak) Speak responses aloud Switch between personality modes (Therapist, Motivator, Funny Friend,etc) Support Languages including: English,Urdu,Punjabi,Korean,Bangoli,Hindi,Pahto,Balochi,French,Arabic,Chinese,Spanish,German

How we built it

Frontend: Streamlit for user interface Voice Input: speech_recognition for capturing user speech Transcription: Whisper (Groq API) NLP & Emotion Tagging: Hugging Face Transformers Text-to-Speech: gTTS Multilingual Support: Google Translate API LLM: Groq API (Mixtral or LLaMA) All code was developed and tested on Google Colab, then deployed to Hugging Face Spaces.

Challenges we ran into

Whisper model import errors due to limited Hugging Face support on some platforms Managing voice-to-text and back-to-voice pipelines smoothly in Colab Ensuring multilingual output while preserving emotional nuance Real-time LLM responses and making them sound human-like Time constraints balancing advanced features like personality switching

Accomplishments that we're proud of

Built a working prototype of a multilingual, emotion-aware chatbot Integrated real-time voice transcription and emotional responses Deployed a fully working demo despite technical hurdles Created something empathetic, inclusive, and comforting

What we learned

Real-time audio processing using Whisper Combining multiple APIs (Groq, Google Translate, gTTS, Hugging Face) Handling language detection and emotional sensitivity in NLP Designing soft, friendly UI/UX for mental health-focused tools Importance of emotional intelligence in AI

What's next for Emoaid:Chatbot for Sensitive Souls

Adding journaling and mental health tracking features Daily motivation messages and check-ins Offline mobile version for private support Deeper emotion classification (e.g., envy, guilt, grief) AI-generated poetry or quotes for healing Integration with professional mental health resource

Built With

  • google-collab
  • google-translate-api
  • groq
  • groq-cloud-api
  • gtts
  • html/css
  • hugging-face-inference-api
  • hugging-face-spaces
  • huggingfacetransformer
  • javascript
  • python
  • speechrecognition
  • streamlit
  • whisper
Share this project:

Updates