Leads the team this role joins
In-house, owns the opening
Director or head of the group
From your own Gmail and LinkedIn.
About the role
This is a mid-level contract research role for someone who builds and tests text-to-speech models, open to candidates anywhere. You would work on expressive voice generation, improving model quality and inference speed, and moving the best results toward production. The role sits within a team of three TTS researchers and works with engineering and hardware teams on deployment. The hiring company is a B2B esports betting data and risk provider.
What you’ll do
- Research, train and evaluate text-to-speech models.
- Experiment with architectures and training data to improve speech quality and speed.
- Move the strongest model results into production.
- Track speech synthesis research and propose improvements based on it.
- Collaborate with the TTS team and with engineering and hardware teams on deployment.
What they’re looking for
- Experience training text-to-speech or voice-cloning models
- Working knowledge of transformers, diffusion models and GANs
- Understanding of speech and audio processing, including sampling, spectrograms and vocoders
- Proficiency in Python, PyTorch and Hugging Face Transformers
- Strong machine learning fundamentals and the ability to read and implement research papers
Nice to have
- Familiarity with GPT-based and flow-matching speech models, Vevo, StyleTTS, IndexTTS or MaskGCT
- Open-source AI contributions or speech research publications
- Familiarity with AWS or similar compute clusters
What’s on offer
- Contract engagement, remote from anywhere
Skills
Questions about this role
Is this role remote?
Yes. The listing says anywhere, as a contract position.
Do I need speech experience?
Yes. Experience training text-to-speech or voice-cloning models is required, along with speech and audio processing knowledge.
What tools should I know?
Python, PyTorch and Hugging Face Transformers are required.