<p style="font-family:"><b><strong style="color:rgb(245,160,32);font-size:10pt;white-space:pre-wrap;">ABOUT PERLE</strong></b></p><p style="font-family:">Perle is an AI infrastructure company building expert-driven training data, evaluation systems, and applied AI products for the world's leading labs and enterprises. Our partners include xAI, Samsung, ELM, and Unisys. Headquartered in San Francisco with experts across more than forty markets, we specialize in the work that requires real human judgment: domain expertise, linguistic nuance, and cultural fidelity that generic data vendors cannot deliver.</p><p style="font-family:"><b><strong style="color:rgb(26,26,26);font-size:11pt;white-space:pre-wrap;">Apply here: </strong></b><a href="https://winnow.perle.ai/jobs/76592750-ce6a-4a73-94c5-8a887781c5d6" class="css-173makr-linkStyle" style="color:rgb(30,74,169);cursor:pointer;" rel="nofollow ugc noopener noreferrer" target="_blank"><u>https://winnow.perle.ai/jobs/76592750-ce6a-4a73-94c5-8a887781c5d6</u></a></p><p style="font-family:"><b><strong style="color:rgb(245,160,32);font-size:10pt;white-space:pre-wrap;">THE ROLE</strong></b></p><p style="font-family:">We are hiring expert Croatian transcribers to convert audio recordings of native speakers into precise, conventionally formatted written transcripts. The work powers speech recognition, dialogue, and evaluation systems for major AI labs. Audio ranges from clean studio recordings to natural conversational speech in real-world acoustic conditions.</p><p style="font-family:">You will be working at the level of expert linguistic judgment. This is not data entry. Decisions about orthography, code-switching, speaker attribution, disfluencies, and non-speech events shape the quality of every model trained on your output.</p><p style="font-family:"><b><strong style="color:rgb(245,160,32);font-size:10pt;white-space:pre-wrap;">DIALECT AND SCOPE</strong></b></p><p style="font-family:">Standard Croatian and its three major dialect groups (Stokavian, Kajkavian, Chakavian). Project audio reflects regional speech with awareness of distinctions from Serbian, Bosnian, and Montenegrin.</p><p style="font-family:"><b><strong style="color:rgb(26,26,26);font-size:11pt;white-space:pre-wrap;">Primary speaker regions: </strong></b>Zagreb, Split, Rijeka, Osijek, Zadar, Dubrovnik.</p><p style="font-family:"><b><strong style="color:rgb(245,160,32);font-size:10pt;white-space:pre-wrap;">WHAT YOU WILL DO</strong></b></p><ul data-pattern="discCircleSquare" data-depth="1" style="font-family:"><li style="color:rgb(245,160,32);font-size:11pt;margin:3px 0px;letter-spacing:0.25px;line-height:1.6;">Transcribe audio recordings into accurate, time-aligned written text following Perle's project-specific style guide</li><li style="color:rgb(245,160,32);font-size:11pt;margin:3px 0px;letter-spacing:0.25px;line-height:1.6;">Apply correct conventions for speaker identification, overlapping speech, disfluencies, fillers, false starts, and n