Senior Engineer — ASR / TTS / Speech LLM (Training + Eval + Integration)

Contribute to training, evaluation, and integration of speech models into OutcomesAI’s voice intelligence stack.You’ll work closely with the Tech Lead to develop datasets, fine-tune models, and benchmark performance across domains (RPM, Triage).

What You’ll Do

• Prepare and maintain synthetic and real trainingdatasets.STT/TTS/Speech LLM model training: from model selection → fine-tuning → deployment.Build evaluation for clinical applications (RPM, Triage, inbound/outbound).
• Build scripts for data selection, augmentation (noise, codec, jitter), and corpus curation.
• Fine-tune models using CTC/RNN-T or adapter-based recipes on multi-GPU systems.
• Implement evaluation pipelines to measure WER, entity F1, and latency; automate MLflow logging.
• Experiment with bias-aware training and context list conditioning.
• Collaborate with backend and DevOps teams to integrate trained models into inference stacks.
• Support creation of context biasing APIs and LM rescoring paths.
• Assist in maintaining benchmarks versus commercial baselines (Deepgram, Whisper, etc.).

Desired Skills

• Strong programming in Python (PyTorch, Hugging Face, NeMo, ESPnet).
• Practical experience in audio data processing, augmentation, and ASR fine-tuning.Training: SpecAugment , speed perturb, noise/RIRs, codec+PLC+jitter sims for PSTN/ WebRTC.Streaming ASR: Transducer/zipformer with chunked attention, frame-sync beam search, endpointing (VAD-EOU) tuning.Context biasing: WFST boosts + neural re-scoring; patient/name dictionaries; session-aware bias refresh.
• Familiarity with LoRA/adapters, distributed training, mixed precision.
• Proficiency with evaluation frameworks WER/sWER, Entity-F1, DER/JER, MOSNet/BVCC (TTS), PESQ/STOI (telephony), RTF/latency at P95/P99. and MLflowlogging.Frameworks: Espnet, Speech brain, Nemo, Kaldi/K2, Livekit, Pipecat, Diffy
• Understanding of telephony speech characteristics, accents, and distortions.
• Collaborative mindset for cross-functional work with ML-ops and QA.

Qualifications

• B.Tech / M.Tech / M.S. in Computer Science, AI, or related field.
• 4-7 years in applied ML; ≥2 years focused on speech recognition or synthesis.
• Experience with model deployment workflows preferred.