Lead ML Data Annotator — Speech & Dialectical AI

Role overview Lead the preparation and quality review of speech data used to build multilingual speech recognition systems and voice-enabled AI. The work combines precise phonetic segmentation, word-level timing, and speaker-aware conversation labeling across regional dialects.

Responsibilities - Curate speech datasets covering major regional dialects, aligning words to audio with millisecond-level timing. - Label conversations to capture multiple speakers and their turns. - Lead acoustic quality reviews, identifying noise, reverberation, and unclear phonetic content. - Create evaluation guidelines for multilingual voice agents and speech-to-text models.

Requirements - Experience with audio labeling, phonetic annotation, or linguistics. - Familiarity with Praat, Audacity, or comparable acoustic annotation tools. - Fluency in English and at least one regional Indian language. - Ability to guide annotation quality and review work.

Benefits and work setup - Remote role based in India, with flexible working hours. - Annual performance bonus and health coverage. - Opportunity to contribute to multilingual speech models.