Contact person: AI/ML Engineer for Lifelike Interactive Avatar System

Phone:Show

Email:Show

Location: Salem, United States

Budget: Recommended by industry experts

Time to start: As soon as possible

Project description:
"We are seeking a highly skilled AI/ML Engineer with expertise in real-time avatar generation, speech technologies, and conversational AI to develop a next-generation interactive avatar system. The project involves building a lifelike AI avatar capable of real-time lip-sync, expressions, and speech-driven interactions using Amazon Web Services (AWS) technologies.

Key Responsibilities:

1. Avatar Creation & Customization
2. Develop an AI-powered avatar generation system where users can upload an image to create a customized avatar.
3. Ensure the avatar replies with perfect lip-sync, eye blinks, facial expressions, head nods, and natural body movements (hands, shoulders).
4. Enable post-upload customization features for users.
5. The avatar should look like the image uploaded by the user, later customisable by the user.

Speech-to-Text Processing:
- Implement Amazon Transcribe for real-time speech capture and transcription.

Conversational AI (LLM Integration):
- Integrate Amazon Bedrock to generate relevant, contextual responses to user queries.

Speech Synthesis:
- Use Amazon Polly to convert generated responses into natural, human-like speech.
- Synchronize speech with avatar lip-sync and expressions (both head and torso).

Real-Time Interaction
- Architect and optimize the system for low-latency, real-time avatar interactions.
- Ensure smooth synchronization across speech recognition, response generation, and avatar rendering.

Required Skills & Qualifications

1. Strong expertise in AI-driven avatar/character animation, real-time rendering, and facial expression/lip-sync modeling.
2. Experience with speech-to-text (STT) and text-to-speech (TTS) technologies, ideally with Amazon Transcribe & Polly.
3. Hands-on experience with LLM integration (Amazon Bedrock or equivalent frameworks).
4. Proficiency in computer vision and deep learning (GANs, diffusion models, or other avatar generation frameworks).
5. Familiarity with avatar/virtual human APIs (e.g., D-ID, Synthesia, HeyGen, Ready Player Me, or similar).
[login to view URL] background in real-time communication systems (WebRTC, sockets, or similar).
7. Familiarity with AWS cloud services for scalable deployment.
8. Strong problem-solving skills, ability to work in an agile setup, and deliver within deadlines." (client-provided description)