Record high-quality scripted English voice samples to train and evaluate text-to-speech models. Remote contract, part-time (20+ hrs/week), requires a quiet studio, professional recording gear, and experience performing long-form audio.
Audio & Speech
100% Remote
Worldwide
Eligibility
Entry
Experience
Jul 18, 2026
Posted
Open worldwide
Interested in this role?
Create a free OpenTrain account and apply in minutes.
OpenTrain is the #1 platform for people building careers in AI training and data labeling. We help skilled contractors find and manage AI training projects, build a unified portfolio of work, and grow a durable freelance career in an industry that’s shaping how AI systems behave.
Why AI training matters
AI systems learn from human examples: voice samples, transcriptions, ratings, and corrections. Contributors provide the real-world recordings and judgments that let text-to-speech and speech models sound natural, consistent, and useful. This work is remote, flexible, and a direct way to influence how cutting-edge systems speak and communicate.
Role overview
OpenTrain AI is recruiting voice actors, narrators, podcasters, broadcasters, commentators, and long-form audio performers to record scripted voice material for an AI text-to-speech (TTS) voice capture project. You will record using an assigned voice persona and deliver clean, dry WAV files suitable for model training and technical review.
Work type: contractor, part-time; minimum 20+ hours/week.
Language: English (recordings in en).
Worldwide contributors accepted; all must be 18 years or older.
Project tasks include audio data collection and evaluation/rating of performance quality.
What you'll do
Perform and submit high-quality, scripted recordings following a specific voice persona and style guide. Maintain consistent delivery across takes and follow all naming, formatting, and submission rules. Participate in re-records when technical or performance issues are found.
Record scripted voice samples using an assigned persona, matching tone, pace, accent, and emotion.
Deliver raw, unprocessed WAV files (no baked-in noise reduction, compression, EQ, reverb, limiter, or AI enhancement unless explicitly approved).
Follow detailed recording instructions, file naming, and upload procedures.
Respond to feedback and complete re-records or corrections as requested.
Requirements
You must be able to produce consistent, professional-sounding recordings and follow technical guidelines precisely. The project requires real recording experience and reliable home-studio conditions.
Proven experience as a voice actor, narrator, podcaster, broadcaster, commentator, or other long-form audio performer.
Access to a quiet, controlled recording environment with no audible background noise.
Professional-quality microphone, audio interface, pop filter or windscreen, and closed-back headphones.
Ability to record and submit raw, dry, unprocessed WAV audio.
Strong vocal control — maintain consistent accent, tone, pace, pronunciation, and persona across takes.
Stable internet access for receiving instructions, uploading recordings, and possible remote direction.
All contributors must be 18 years of age or older.
Who should apply
Apply if you are an experienced audio performer who can deliver consistent, repeatable performances in a professional home-studio setup and want remote, part-time contract work shaping TTS models.
Good fit: professional voice actors, audiobook narrators, experienced podcasters, radio broadcasters, and experienced voice-over artists.
Not required: formal linguistics or speech science credentials — performance experience and reliable recordings are the core qualifications.
How it works
OpenTrain AI manages onboarding, project instructions, and quality review. After application you’ll receive detailed style guides, recording instructions, and submission portals. Assignments include both new data collection and evaluation/rating tasks to ensure recordings meet technical and performance standards. This is contract, part-time work; more project details and compensation information will be shared during the application and onboarding process.
Onboarding includes technical checks, sample recordings, and approval before full assignment.
You will submit WAV files via the provided upload workflow and may be asked to re-record clips that fail quality checks.
Employment type: contractor; hours: 20+ per week (part-time).
Project data types: AUDIO; label types: DATA_COLLECTION and EVALUATION_RATING.
Join OpenTrain to record high-quality voice samples for text-to-speech training and evaluation — remote, contract work at $50/hr. Native American English speakers in the US with a professional or near-professional home studio and voice-acting experience are encouraged to apply for 20+ hours/week.
Record high-quality Northern English voice samples for text-to-speech training at $50/hour. Part-time contractor role (20+ hrs/week) requiring a professional home recording setup and experience in voice work.
Record high-quality African American English voice samples for next-generation text-to-speech at $50/hr. Part-time contract work (20+ hrs/week), remote across the US; professional/near-professional recording setup and voice-acting experience required.
