The project goal is to create a voice chatbot, on option A without avatar and on option B with avatar. Please submit estimated time either per option, A or B, or per version (A1, A2, A3, B1, B2, B3) OPTION A — Voice Chatbot (no avatar) Version A1 — Flexible (multiple platforms) Title: Prototype Voice Chatbot (Android Tablet) – STT + LLM + TTS Description: We need a quick prototype (2–3 weeks) of a voice chatbot for Android with the following features: - Speech-to-Text (STT): Google Speech-to-Text or Whisper API (OpenAI). - LLM: OpenAI GPT-4o / GPT-4o-mini, or similar. - Text-to-Speech (TTS): Google Cloud TTS, Azure TTS, or ElevenLabs. - Simple Android app: push-to-talk button, voice reply playback. - Optional backend to handle API calls (Firebase, AWS Lambda, etc.). Deliverables: Functional APK + demo video + short documentation. Timeline: 2–3 weeks Version A2 — Unified (Google Cloud) Title: Prototype Voice Chatbot (Android Tablet) – Google Cloud AI Description: We want a functional prototype of a voice chatbot on Android tablet, based entirely on Google Cloud. - STT: Google Cloud Speech-to-Text - LLM: Gemini 1.5 Flash or Vertex AI (PaLM/Gemini) - TTS: Google Cloud Text-to-Speech - Backend: Firebase Functions - Android App: Flutter or Android Studio Deliverables: APK + demo video + documentation. Timeline: 2–3 weeks Version A3 — Unified (AWS) Title: Prototype Voice Chatbot (Android Tablet) – AWS AI Description: Quick prototype of a voice chatbot for Android, built on Amazon Web Services (AWS). - STT: Amazon Transcribe - LLM: Amazon Bedrock (Claude, Llama 3, etc.) - TTS: Amazon Polly - Backend: AWS Lambda - Android App: Flutter or Kotlin/Java Deliverables: APK + demo video + documentation. Timeline: 2–3 weeks OPTION B — Chatbot with Photorealistic Avatar Version B1 — Flexible (multiple platforms) Title: Prototype Chatbot with Photorealistic Avatar (Android Tablet) Description: We need a quick demo prototype (3–4 weeks) of a chatbot with a photorealistic avatar for Android. - STT: Google Speech-to-Text or Whisper API - LLM: OpenAI GPT-4o / GPT-4o-mini or Gemini - TTS: Google Cloud, Azure, or ElevenLabs - Avatar: Heygen API or D-ID API (preferably with real-time streaming) - Emotion: Azure Cognitive Services or similar - Android App: basic UI with microphone + avatar video Deliverables: APK + demo video + documentation. Timeline: 3–4 weeks Version B2 — Unified (Azure) Title: Prototype Chatbot with Avatar (Android Tablet) – Azure AI + D-ID Description: Prototype of a chatbot with photorealistic avatar, based on Microsoft Azure Cognitive Services. - STT: Azure Speech-to-Text - LLM: Azure OpenAI (GPT-4o / GPT-4o-mini) - TTS: Azure Neural Voices - Avatar: D-ID API (real-time streaming) - Emotion: Azure Sentiment Analysis (basic positive/negative/neutral) - Backend: Azure Functions - Android App: Flutter Deliverables: APK + demo video + documentation. Timeline: 3–4 weeks Version B3 — Unified (AWS) Title: Prototype Chatbot with Avatar (Android Tablet) – AWS AI + D-ID Description: Prototype of a chatbot with photorealistic avatar, built on Amazon Web Services (AWS). - STT: Amazon Transcribe - LLM: Amazon Bedrock (Claude, Llama 3, etc.) - TTS: Amazon Polly - Avatar: D-ID API - Emotion: Amazon Comprehend (basic sentiment analysis) - Backend: AWS Lambda - Android App: Flutter Deliverables: APK + demo video + documentation. Timeline: 3–4 weeks