US English Audio Transcription QA Specialist

About OpenTrain AI

OpenTrain AI is the #1 platform for finding and building careers in AI training and data labeling. We make it easier to find specialized projects, build a lasting portfolio, and grow a freelance career teaching AI.

OpenTrain AI is the hiring and contracting organization for this role. Creating an OpenTrain account is free and gives you access to this project and others across the industry.

Why AI Training Work Matters

AI training (data labeling and human evaluation) is the human side of building modern AI systems. Contributors provide the examples and quality checks that teach models to understand speech, language, images, and more.

This work is 100% remote, often flexible and part-time, and gives you direct influence on how state-of-the-art systems behave. Many projects require no prior experience beyond language fluency and attention to detail; specialist roles pay more for domain expertise.

The Role

As a US English Audio Transcription QA Specialist you will serve as the final quality checkpoint for an evaluation-grade dataset of multi-channel, unscripted group conversations. Your reviews ensure recordings, transcriptions, timestamps, speaker labels, and metadata meet strict technical and accuracy standards.

This is an entry-level contractor position with a time expectation of 20+ hours per week. Work is fully remote and open to worldwide applicants who meet the language and technical requirements below.

• Employment type: Contractor, Part-time.
• Time requirement: 20+ hours/week.
• Language: Native-level US English required.
• Data type: Multi-channel audio, human-verified transcripts, JSON metadata.

What You'll Do

You will perform detailed, end-to-end QA of audio files, transcriptions, and metadata. Reviews are technical and accuracy-focused—you will confirm the dataset meets strict fidelity and verbatim transcription targets and flag issues for correction.

• Evaluate multi-channel recordings for channel isolation, background noise, clipping, low gain, and other fidelity issues.
• Review human-validated transcripts for exceptionally high accuracy and adherence to unnormalized, verbatim transcription rules and word-error-rate targets.
• Validate turn-level and word-level timestamps, speaker identification, and metadata tags (including demographic and contextual markers).
• Read and validate JSON-formatted metadata and timestamp logic for consistency and correctness.
• Audit sessions for personally identifying information, toxic or sensitive content and enforce safety and privacy standards.
• Assign clear pass/fail or agree/disagree statuses and provide detailed, actionable feedback when corrections are needed.

Requirements

You must preserve verbatim transcription rules and demonstrate the listening and analytic skills to QA multi-speaker conversational recordings accurately.

• Native-level US English proficiency and excellent ear for subtle audio issues (background noise, channel bleed, clipping).
• Meticulous attention to detail for verifying word-level timestamps and strict, unnormalized verbatim transcription.
• Ability to assess complex multi-speaker dynamics, including overlaps, false starts, and backchannels.
• Experience validating word-level timestamps and following rigorous timestamping guidelines.

Who Should Apply & How It Works

Ideal applicants include linguists or phonetics experts, language teachers with native US English mastery, and professional transcriptionists experienced in tight timestamping and accuracy targets. The role is entry level but requires domain-relevant listening and transcription QA skills.

OpenTrain AI will manage contracting, instructions, and test tasks. You will complete qualification checks and sample reviews before working live. There is no pay information in this listing; compensation details are provided during the application and onboarding process on the OpenTrain platform.

• Helpful backgrounds: linguistics/phonetics, language teaching, professional transcription.
• Worldwide applicants are welcome if you meet the language and technical requirements.
• To apply: create an OpenTrain account, complete the qualification steps, and submit required sample reviews as part of onboarding.