Note: The job is a remote job and is open to candidates in USA. OVA.Work is seeking a Speech Recognition Engineer to design, develop, and optimize Automatic Speech Recognition (ASR) systems for voice-enabled applications. The role involves building, training, fine-tuning, and deploying speech recognition models with high accuracy and robust performance across diverse languages and acoustic environments. The engineer will collaborate with cross-functional teams and deploy models using MLOps and cloud-native practices.
Responsibilities
- Design, develop, and optimize Automatic Speech Recognition (ASR) models for production applications
- Build end-to-end speech processing pipelines, including audio preprocessing, feature extraction, decoding, and post-processing
- Train, fine-tune, and evaluate speech recognition models using large-scale speech datasets
- Improve recognition accuracy for multilingual, domain-specific, and noisy audio environments
- Develop real-time and batch speech recognition solutions
- Optimize models for latency, throughput, memory efficiency, and inference performance
- Integrate ASR models into voice assistants, conversational AI systems, call center platforms, and enterprise applications
- Develop data pipelines for speech data collection, annotation, augmentation, and quality validation
- Evaluate model performance using industry-standard speech recognition metrics
- Collaborate with NLP Engineers, Machine Learning Engineers, AI Engineers, Data Scientists, and Product teams
- Deploy speech recognition models using MLOps and cloud-native deployment practices
- Monitor production performance and continuously improve model quality
Skills
- Bachelor's or Master's degree in Computer Science, Artificial Intelligence, Machine Learning, Electrical Engineering, Speech Technology, or a related field
- 3+ years of experience in speech recognition, speech processing, machine learning, or AI engineering
- Strong programming skills in Python
- Experience with deep learning frameworks such as PyTorch or TensorFlow
- Solid understanding of digital signal processing (DSP) fundamentals
- Experience with speech processing libraries such as SpeechBrain, ESPnet, Hugging Face Transformers, torchaudio, librosa, or Kaldi
- Experience training and fine-tuning deep learning models
- Familiarity with Linux development environments, Git, and containerization using Docker
- Understanding of cloud platforms such as AWS, Microsoft Azure, or Google Cloud Platform
- Experience with modern ASR architectures such as Whisper, Conformer, wav2vec 2.0, DeepSpeech, or RNN-Transducer (RNN-T)
- Experience deploying speech recognition models using ONNX Runtime, TensorRT, NVIDIA Triton Inference Server, or TorchServe
- Knowledge of multilingual and low-resource language speech recognition
- Experience with streaming speech recognition and real-time inference
- Familiarity with speech enhancement, voice activity detection (VAD), speaker diarization, and keyword spotting
- Experience with MLOps tools such as MLflow, Kubeflow, or cloud AI platforms
- Knowledge of Large Language Models (LLMs) for speech understanding and conversational AI
Company Overview