Portuguese Audio Annotation QA Specialist

About OpenTrain

OpenTrain is the #1 platform for building careers in AI training and data labeling. We help freelancers find projects, consolidate proof-of-work, and grow a lasting portfolio of AI training experience — all in one place.

As the hiring organization, OpenTrain connects contributors directly with high-quality annotation and QA work so you can build reliable, repeatable freelance income in a fast-growing field.

Why AI Training Work Matters

AI training (also called data labeling or annotation) is the human side of building intelligent systems: people prepare and verify the examples models learn from — transcripts, labels, and feedback that shape model behavior.

This kind of work is often 100% remote, flexible, and accessible to entry-level contributors with language ability and attention to detail. It’s a way to work on cutting-edge AI while building a specialized portfolio.

The Role

We are hiring a Portuguese Audio Annotation QA Specialist to review multi-channel conversation recordings and the human-validated transcripts and diarization that accompany them.

You will audit audio fidelity and channel isolation, confirm verbatim transcription and word-level timestamps, validate speaker labels and metadata across overlapping dialogue, and flag PII or other sensitive content.

• Work type: Contractor, part-time.
• Time commitment: 20+ hours per week.
• Location: Remote — worldwide applicants accepted.
• Languages required: Native Portuguese (pt).

What You’ll Do (Key Responsibilities)

Your day-to-day work centers on careful listening, detailed comparison of transcripts to audio, and clear written feedback for annotators. You will examine both audio files and JSON-formatted annotation data.

• Review multi-channel audio for fidelity, quiet recording conditions, and proper channel isolation.
• Validate human-checked transcripts for strict verbatim accuracy and low error rates at the word level.
• Confirm turn-level and word-level timestamps, speaker labels, and metadata in overlapping, multi-speaker dialogues.
• Audit sessions for PII, toxic content, and other sensitive material and flag items per guidelines.
• Provide actionable, concise comments when annotations do not meet quality standards.

Requirements

This role requires specific audio and language skills rather than previous industry seniority; it is suitable for entry-level contributors who meet the technical requirements below.

• Native proficiency in Portuguese required.
• Strong ear for subtle audio defects such as bleed, clipping, and low gain.
• Proven attention to word-level transcription and timestamp accuracy.
• Comfort assessing complex multi-speaker conversations, overlaps, and speaker diarization.
• Familiarity reading and commenting on JSON-formatted annotation output.
• Ability to reliably identify PII and potentially harmful or toxic content in recordings.

Who Should Apply

Apply if you are a native Portuguese speaker who enjoys careful listening, detail-oriented review work, and giving constructive feedback to improve annotation quality.

This position fits freelancers looking for steady part-time contract work (20+ hours/week) and contributors who want to build verifiable QA experience in AI training.

How the Process Works

Create or use your OpenTrain profile to apply. You will receive project instructions, QA guidelines, and sample tasks to ensure consistent reviews. Compensation and project-specific details are provided within each assignment; OpenTrain manages contracts and project access.

Expect to work remotely, review audio and JSON files in your browser or chosen tools, and submit clear annotations and feedback according to provided rubrics.