Internship Opportunity: AI/ML Engineer (Speech & MLOps)

Location

Engagement

Compensation

Preference

We’re looking for a skilled AI/ML Engineer with Speech Recognition expertise and MLOps experience to help build the core intelligence of this life-changing product.

Responsibilities

• Design and implement personalized speech correction systems using user-specific data.

• Analyze low-quality audio and implement noise reduction, voice smoothing, and fluency correction techniques.

• Fine-tune open-source ASR models like Whisper, DeepSpeech, or Wav2Vec2 for impaired speech in Marathi.

• Train text-to-speech (TTS) models using tools like Coqui TTS, Google TTS APIs, or custom vocoders.

• Perform data cleaning, preprocessing, augmentation (speech).

• Develop reproducible training pipelines using PyTorch, TensorFlow, or scikit-learn.

• Conduct model hyperparameter tuning.

• Collaborate closely with product, design, and mobile development teams to integrate models effectively. • Ensure model generalization and robustness through continuous improvement and testing. 

Requirements

• Proven experience in Speech Recognition (ASR) and/or Speech Synthesis (TTS).

• Hands-on with PyTorch, TensorFlow, HuggingFace Transformers.

• Familiarity with Librosa, Waveform manipulation, and audio signal processing.

• Good command over Python, Git, and training pipeline setup.

• Knowledge of language models, accent correction, or phoneme-level tuning.

• Experience with custom dataset creation, augmentation, and weak supervision.

• Experience working with Indian regional languages (plus point).

Work Culture & Deliverables

• You’ll be working directly with the founder (tech/product-led).

• Weekly deliverables with clear outcomes and model benchmarks.

• Ethical focus: user privacy, offline-first experience, and real-world usability