Skip to content
RR-GPU-Accelerated-Speech-Capabilities​ - LuxAI S.A.

GPU-Accelerated Speech Capabilities

QTrobot’s advanced microphone array and AI speech capabilities enable dynamic and emotive multi-lingual conversations.

Discover unparalleled speech recognition and translation accuracy with Nvidia Riva, experience the emotional depth of Acapella’s text-to-speech voices, and explore innovative Generative AI voices like Bark – all seamlessly operating in real-time on the device.

ReSpeaker V2

  • 4 high performance digital mic
  • Far-field Voice Capture
  • Voice Activity Detection
  • Direction of Arrival
  • Noise Suppression
  • Acoustic Echo Cancellation

Nvidia® Riva

  • GPU-accelerated multilingual speech and translation
  • AI software development kit for automatic speech recognition
  • Text-to-speech and neural machine translation


Emotional text to speech  voices for over  30  languages and 200 voices,  including unique and  genuinely rich repertoire of  truly lifelike children  voices


Fully generative text-to-audio model based on a GPT style architecture  like  AudioLM  and  Vall-E. It can  generalize to arbitrary instructions  beyond speech such as music lyrics and sound effects