OpenAI Hints at Alexa and Siri Rival With New ‘Voice Engine’ Trademark Filing
OpenAI has filed a trademark application with the U.S. Patent and Trademark Office for products related to “digital voice assistants” and a “voice engine.” The trademark filing, suggests that OpenAI has plans for more voice-centric generative AI assistants, beyond the current speech recognition and text-to-speech capabilities provided through ChatGPT and related products.
GPT-Voice
Specifically, the filing covers software for building digital voice assistants, voice command processing, audio generation from text prompts, multilingual speech recognition and translation, and voice service delivery, among other areas. OpenAI already offers speech and audio technologies through APIs, such as its text-to-speech API and Whisper speech recognition model. However, the company doesn’t currently have a specific virtual voice assistant product for consumers.
ChatGPT can also vocalize its responses using any of five different voices through the Read Aloud feature that came out recently. The feature produces an audio performance of ChatGPT’s answers in 37 languages, automatically detecting which language is being used and pronouncing the words accurately. ChatGPT also has Whisper, which enables the chatbot to understand a user’s speech and respond vocally. However, that was a purely audio experience, with no written text. Read Aloud goes one step further by reciting written responses. OpenAI has suggested the spoken format could be beneficial for those multitasking or who need their hands free.
OpenAI has been busy at the USPTO in recent months. The agency recently denied OpenAI’s attempt to trademark the term “GPT,” as descriptive of the technology itself and, therefore, not something that can be trademarked. That said, trademarking GPT-5 and even later potential iterations GPT-6 and GPT-7 have not faced the same obstacles. GPT-5 is expected to roll out some time this summer, according to recent reports.
Follow @voicebotai Follow @erichschwartz
OpenAI’s Whisper ASR is Transcribing Apollo 13’s 7,200 Hours of Mission Control Chatter
