Lead Speech Data Linguist: Standards & QA at Scale

Innodata Inc.; United States; Remote; USD 160,000 - 185,000; Full time

Benefits offered by this job

Remote-friendly policy

Job summary

Innodata Inc. in the United States seeks a principal-level linguist to own linguistic standards for segmentation and transcription across languages and use cases. You will define conventions, ensure quality, and guide end-to-end work flows.

You will mentor staff, collaborate with researchers, and shape human-in-the-loop strategies as models evolve, aligning linguistic objectives with ASR downstream tasks.

Qualifications

• Substantial industry experience in transcription/segmentation and speech-data quality.
• Bachelor's in linguistics/phonetics/computational linguistics; advanced degree preferred.
• Fluency in IPA and phonetic transcription; experience with acoustic analysis.

Responsibilities

• Define transcription and segmentation standards, style guides, and annotation conventions.
• Establish and run quality frameworks: rubrics, error taxonomies, adjudication, and inter-annotator agreement.
• Own end-to-end quality lifecycle for transcription/segmentation deliverables; ensure scalable QA and reporting.
• Design human-in-the-loop workflows to maximize value as ASR quality improves.
• Mentor transcribers and reviewers; build onboarding/calibration processes.

Skills

Linguistic standards

IPA phonetics

Phonetic transcription

Acoustic analysis

Praat scripting

Python scripting

Inter-annotator agreement

Multilingual data

Education

Bachelor's in linguistics/phonetics/computational linguistics

Tools

ELAN

Montreal Forced Aligner

Whisper/ASR engines

Praat

Job description

Innodata Inc. in the United States seeks a principal-level linguist to own linguistic standards for segmentation and transcription across languages and use cases. You will define conventions, ensure quality, and guide end-to-end work flows.

You will mentor staff, collaborate with researchers, and shape human-in-the-loop strategies as models evolve, aligning linguistic objectives with ASR downstream tasks.