Job Description
We are hiring English (UK) voice actors to perform multi-character dialogue recordings to help train internal Text-to-Speech (TTS) models. You'll be voicing fictional characters with varying styles, accents, and personalities. This will be a SINGLE person recording project, but you will be expected to perform different voices.
_ This is non-broadcast, non-commercial work for internal research and development purposes only. Your voice will not be used in public-facing or paid media._
Project Scope
- Record 3 hours of audio performing dialogues between 2+ characters
- Adjust voice, tone, and accent based on character descriptions
- Ensure clear differentiation between characters
- Deliver RAW, high-quality WAV files according to project specs
- Complete up to 2 rounds of revisions (for talent errors)
Audition & Selection Process
- Submit an audition using only ONE of the sample scripts attached below.
- _ IMPORTANT_ : If you're name starts with A-K please select ONE of the first three scripts. If it starts with L-Z, please select ONE of the second three scripts for your audition. (These scripts will be provided once you are done submitting your proposal in the Upwork platform)
- Please submit auditions with the following audio specs:
- Format: 24-bit / 48 kHz / mono / unprocessed .wav
- Minimum: -60 dB noise floor
- No plugins, compression, or post-processing
- Clean recordings with minimal background noise, breaths, and clicks
- The audio MUST be narrated by a human. AI is NOT allowed.
- After you are selected, the three hours of audio will need to be delivered within one week.