Translation earbuds employ a three-stage process to convert speech between languages: capturing audio through multiple microphones with noise-separation technology, processing the audio through speech recognition and mac…
#Audio & Voice AI
Real-time translation earbuds continue advancing through multi-stage processing architectures that combine noise filtering, speech recognition, machine translation, and audio synthesis, with manufacturers diverging on whether computation happens locally or via cloud services. Meanwhile, creative applications of audio AI are emerging, including experimental music devices that deliberately exploit AI artifacts rather than polish them away. The sector faces mounting legal challenges, as major record labels escalate copyright disputes against AI music platforms, alleging sophisticated training practices designed to circumvent licensing requirements. These developments highlight the tension between audio AI's expanding capabilities and unresolved questions about training data legitimacy.
Thoughtful Things launched a Kickstarter for Engram, a sampler and groovebox that uses locally run AI to transform incoming audio and generate unusual sounds. The device is not meant to produce polished songs, but to hel…
Sony and Universal Music Group filed another copyright suit against Suno, arguing that its v6 model was trained on outputs from earlier models that used unlicensed music. The labels call this practice 'model laundering' …
Google's AI language efforts now cover more than 300 languages, aiming to understand how people actually speak, including tone and slang. The company is focusing on cultural nuance and working with local communities to e…