Top Tech Jobs & Startup Jobs

Transcribe Hindi audio into accurate, well-formatted written text across single- and multi-speaker recordings, podcasts, and conversations. Identify speakers, follow project-specific formatting and quality guidelines, and perform annotation/tagging tasks when required. Work remotely via a secure platform with ongoing support and flexible hours.

Transcribe Japanese audio into accurate, well-formatted text across single- and multi-speaker recordings and podcasts. Identify speakers when needed, follow project guidelines, and perform annotation/labeling tasks as required to support audio processing and ML training datasets.

Flexible remote project collecting and validating Ukrainian speech and text data for an AI dataset. Tasks include recording short sentences at 8 kHz, transcribing audio, cleaning and organizing text, and optionally reviewing others' work. Over 40 hours available; project runs spring to early summer. Requires reading guidelines and passing a short quiz. Paid hourly.

Compare original texts with Traditional Chinese (Taiwan) machine translations to judge accuracy, grammar, and naturalness. Provide simple quality judgments (Pass/Fail or Yes/No) following guidelines. Weekly engagement ~1–5 hours for one year; no rewriting required.

Contribute to an AI project by recording Portuguese (EU) voice samples, transcribing audio, cleaning and organizing text, and optionally reviewing others' work. Flexible, online, paid hourly; requires fluent Portuguese (EU), fluent English, prior audio/transcription/annotation experience, attention to detail, and a quiet recording space.

Freelance remote role to record ~1.5-hour scripted conversational sessions in Russian and English with natural code-switching. Complete a short pre-vet recording, be matched with a partner, and record sessions on the Aurora Studio platform.

Compare source texts with Hebrew machine translations to judge accuracy, grammar, and naturalness. Make simple pass/fail quality assessments following provided guidelines. Weekly commitment 1–5 hours for a year; no rewriting required.

Transcribe German audio into accurate, well-formatted text for single- and multi-speaker recordings, podcasts, and conversations. Identify speakers where applicable, follow project-specific guidelines, and perform annotation/labeling tasks as needed. Maintain high accuracy and meet scheduled shift availability.

Record a ~1.5-hour scripted conversational session with another speaker, using natural code-switching between Saudi Arabian Arabic and English. Complete a short pre-vetting test on the Aurora Studio Platform, then participate in scheduled remote recordings in a quiet environment with reliable internet.

Transcribe Danish audio into accurate, well-formatted text; identify speakers; follow project-specific guidelines; optionally perform annotation and labeling to support training datasets.