Urdu Voice Artist (Male or Female)

Urdu Voice Artist — Voice Corpus for AI Research (Male or Female) Awaz Media is building a studio-grade Urdu voice corpus for AI and TTS (text-to-speech) research. We're hiring voice artists to join our recording and annotation team. We're specifically recruiting both male and female voices, because a corpus trained on only one gender produces a TTS system that only sounds like one gender. What you'll do Primary: read Urdu scripts we provide into a microphone in a clear, sustained, consistent voice. Recording sessions happen at a location we arrange; you don't need your own equipment for the sessions themselves. Secondary: after each recording session, listen back to your own recordings in a web tool (we train you on it) and transcribe them — verbatim first, then a clean normalized version suitable for TTS training. Tag quality flags (background noise, misreads, code-switches). Because you're transcribing your own voice, this is faster than transcribing a stranger — but it needs the same discipline. The corpus is only as good as the labels. Required Native Urdu with a clear, warm reading voice. We welcome every clearly articulated regional register — Karachi, Lahore, Rawalpindi/Pothohari, Peshawar, Multan, Hazara, Kashmir, Hyderabad, or wherever you grew up. We're building a dataset that represents how Urdu is actually spoken, not one single "correct" accent. Fluent reading of long Nastaliq passages, including Arabic and Persian loanwords without stumbling. Working English — our training materials and the labeling tool are in English, and the recordings themselves will contain code-switched English words. Ability to type in Nastaliq (Phonetic or InPage layout). Focused stamina for 4-6 hours of recording and transcription a day. Your own laptop, headphones, and stable internet. Preferred Prior voiceover, radio, news reading, naat recitation, tilawat, or audiobook narration experience. Theatre or acting background. Urdu literature coursework at university level. Compensation PKR 80,000/month full-time. Paid 1-week trial on real recording and transcription work before confirming the role. Recording sessions on-site at a location we arrange (city shared with shortlisted candidates); transcription can be done from home. Flexible hours around weekly targets. To apply Fill out the form below. It takes about 5-7 minutes. You'll upload your CV and record two or three short voice samples right in the browser (no software needed — just allow microphone access when prompted). One sample is a prescribed passage every applicant reads; one is spontaneous speech on a topic of your choice; the third (a longer prescribed passage) is optional but recommended. All recording happens privately on your device and only uploads when you submit. Shortlisted candidates will hear back within 3 days.