TTS/STT Expert (m/f/d) — View the role in Essen

As a TTS/STT Expert, you are responsible for the further development of our text-to-speech and speech-to-text systems. You work to noticeably improve the naturalness and emotional quality of the generated speech for our avatars, at the forefront of modern speech technology – from Whisper and CosyVoice to F5-TTS. You are part of an international team that closely connects research and production.

• Development and optimisation of our speech synthesis and speech recognition systems
• Improving natural language processing for smoother dialogues
• Optimising latency and quality in real-time speech processing
• Integrating existing speech APIs and developing in-house solutions

Your profile

• Master's degree or PhD in speech technology or computational linguistics
• Experience with deep learning for speech synthesis and recognition
• Knowledge of Python and C++
• Familiarity with modern TTS/STT models and APIs (e.g. Whisper, CosyVoice, F5-TTS or Google/Amazon Speech)
• Research experience in TTS/STT is an advantage

What we offer

• Work on one of the most visible quality features of our avatars: How natural and emotional they sound depends directly on your work
• Real-world performance instead of lab conditions: Your optimisations directly affect latency and the experience of real users
• Access to state-of-the-art hardware and room for your own research approaches instead of simply implementing predefined models
• International team focused on speech technology instead of working alone
• Free choice between remote work and our office in Essen
• A short, straightforward application process: initial call, video interview, team meeting
• Annual salary of 50.000–70.000 €, depending on experience

#J-18808-Ljbffr