Research Intern - Audio-Visual VoiceAI (Open Source)

Job Description

Research Intern – Audio-Visual VoiceAI (Open Source)

We’re looking for a Research Intern to join WhissleAI and help advance our open-source work at the intersection of speech, vision, and structured understanding — inspired by projects like

• https://music.whissle.ai/
• advanced speech recognition asr.whissle.ai
• and recent multi-modal alignment research (example: https://aclanthology.org/2025.emnlp-main.845.pdf)

You’ll work on developing audio-visual foundation models that connect voice, context, and environment — enabling systems that can listen, see, and act coherently in real time. Most of this work is open-source and contributes directly to the broader research community.

Ideal candidate

• Undergrad, Master’s, or PhD student in CS, AI, or related field
• Prior research experience (conference/workshop publications a plus)
• Strong background in one or more of: multimodal learning, audio-visual representation learning, speech modeling, or self-supervised methods
• Experience with PyTorch, Hugging Face, or similar frameworks

What you’ll do

• Prototype and evaluate audio-visual alignment models
• Extend our open-source ASR and meta-speech pipelines
• Collaborate on papers, demos, and real-time VoiceAI applications

Location: Remote

• Type: Paid internship / research collaboration