Job Requirements
3 - 5 years of experience
Skills
Job description for Voice/Speech Research Engineer at Sentient.io
DUTIES & RESPONSIBILITIES
- Benchmark and validate predictive speech models for technologies such as, automatic speech recognition, speaker recognition, text-to-speech etc.
- Source for different types of speech datasets including data collection and create new ones as needed.
- Incorporate a wide variety of statistical and machine learning techniques such as deep learning, classification and regression models etc.
- Leverage cutting edge cloud platforms such as GCP to create automated training and testing pipelines
ARCHETYPES OF THE SUCCESSFUL CANDIDATE
- Implement and validate predictive speech models for technologies such as, automatic speech recognition, speaker recognition, text-to-speech etc.
- Two to five years experience in one or more of the following areas: speech processing, automatic speech recognition, speaker recognition, text-to-speech preferred.
- Programming expertise including Shell, Perl, Python with the ability to use Linux command line tools.
- Experience working with open source speech libraries, like Kaldi, Alize, DeepSpeech etc.
- Experience with working on cloud computing platforms, such as, GCP, AWS etc.
- A clear communicator with the ability to deliver timely.
- Fresh graduates with relevant experience are welcome to apply.
KEY REQUIREMENTS
- BS/MS in Computer Science, Signal Processing, Statistics, Mathematics, etc
- Minimum 2 to 5 years experience in the areas of speech signal processing, speech processing, machine learning preferred, and fresh graduates with good honours may be considered
- Strong grasp of speech processing and machine learning techniques
TECHNICAL REQUIREMENTS
- Programming: Python, Shell, Perl
- Libraries: Kaldi, librosa, pyannote, pytorch, tensorflow
- End-to-end pipeline development and development in Kubeflow
- Basic knowledge of software engineering (git, devops processes, etc.) and REST APIs would be preferred