AI Data Annotation Intern (Transcription) - Remote, Part-Time
Remote · Part Time internship
15 applicants
- Stipend
- Stipend: INR 25,000 – INR 30,000 / month
- Duration
- 3 months
- Start
- Immediately
- Openings
- 30
Candidates eligible for this internship must have the required skills and interests, be able to work from home, start between September 25 and October 30, 2026, and commit to the full 3-month duration.
- Work mode
- Work from home
- Resume
- Required to apply
About the internship
Internship Overview
This internship involves carefully listening to audio clips and reviewing their corresponding transcripts according to specific annotation guidelines. Tasks include writing all text in lowercase, tagging filled pauses and backchannels appropriately, identifying transcription mistakes, and ensuring accurate word-for-word transcription. The role demands high attention to detail and is best suited for individuals who enjoy focused, quiet work and have strong language skills.
Key Responsibilities
- Listen attentively to audio files and verify transcripts against actual spoken words.
- Identify and correct errors such as missing or extra words in transcripts.
- Consistently apply detailed annotation instructions across large quantities of data.
- Flag unclear audio segments, unusual cases, or content beyond the scope of the guidelines.
Candidate Requirements
- Availability for a remote internship starting between September 25, 2026, and October 30, 2026, lasting 3 months.
- Strong English comprehension and writing skills to understand and execute detailed guidelines.
- Access to a laptop or desktop computer (mobile devices are not suitable for this role).
- Use of headphones and a quiet environment for precise listening.
Perks
- Flexible working hours allowing interns to choose their schedule.
About the Company
Zyno AI India Private Limited specializes in developing AI products aimed at simplifying content creation. Their flagship product, Zyka AI, assists creators in producing high-quality videos, images, and audio efficiently.