Looking to solve complex ASR challenges at scale?

You'll be joining an established conversational AI company with proprietary in-house speech models that process billions of interactions annually, where your ASR expertise will directly impact real-world customer experiences.

You'll be tackling demanding ASR problems in production environments: streaming speech recognition in noisy conditions, robust accent handling and maintaining high performance at scale.

Working across key production environments, you'll enhance speech capabilities and bring new features to production that push the boundaries of what's possible in challenging acoustic environments.

Your focus

• Maintain and iteratively improve existing ASR technology while introducing cutting-edge enhancements
• Work end-to-end across speech processing components: speech enhancement, VAD, diarisation, and ASR (AM/LM modelling, ASR biasing)
• Build streaming ASR systems optimised for challenging acoustic environments
• Implement emotion detection and acoustic condition classification capabilities
• Run extensive experiments to advance activity detection and speech processing performance

What you'll bring

• Strong background in ASR model development and deployment
• Hands-on experience with SOTA speech toolkits (Kaldi, K2, NVIDIA NeMo, Parakeet)
• Proven streaming ASR experience in production environments

This is a fully remote role – must be close to EU timezone. 

Ready to make your mark on speech tech that millions rely on daily? Apply today.

The future of communication should be bias and barrier-free. That's the vision behind this well-funded start-up pioneering real-time speech algorithms.
 
You'll join a research team on tech that is the first of its kind, improving how we communicate in the real world. By offering clear, natural-sounding conversations regardless of accent or environment. Their ground-breaking technology is already providing impressive results, so it's no wonder they're growing x4 annually.
 
They are creating the whitespace in speech research, and you'll play a key role.
 
As an Senior ML Scientist, you'll work within a talented R&D team advancing core speech algorithms and audio AI models.
 
The role
 
- Contribute to cutting-edge R&D advancing core speech algorithms and Generative Audio models. Continually push boundaries to the next level.
- Tackle unsolved problems in Generative Speech and Audio such as preserving naturalness and performance in noisy environments.
- End-to-end ownership of models, from data collection to training on the cloud.
- Develop novel architectures balancing cutting-edge performance with real-time efficiency & low-latency
- Collaborate with top scientists in this field 
 
You'll have
 
- 4+ years of industry experience developing and implementing either of the following: TTS, Voice Conversion/Cloning, Speech Synthesis, Speech Translation, Accent Translation, Speech Restoration
- Proven background contributing to well-known research publications and/or products in these areas
- PhD or degree in Computer Science, ML, or related field. 
- Proven experience with PyTorch, TensorFlow and modern DL techniques such as GANs, VAEs, diffusion or flow models, etc.
- Familiar with cloud-based technologies and production environments
 
What you'll get in return
 
- Benefits include a competitive salary, share options, unlimited PTO health coverage, and a VPO plan.
- Contributing to whitespace in speech technology research, you'll have control over the direction of your work with no friction whatsoever. You're the expert after all.
 
If you're looking to make an impact, there are few better places to do it. Your work here has the power to improve communication, eliminate confusion, and create a more connected world.
 
If you want the freedom to shape the future of speech AI, apply now.