Sam Altman may be striving to tackle Apple’s Siri and Amazon’s Alex voice assistants.
OpenAI has filed a trademark software to construct “digital voice assistants” and a “voice engine”, which alerts that it might be set to launch a brand new product.
The applying was filed with the US Patent and Trademark Workplace on Tuesday, a day after an interview with Sam Altman was broadcast during which he mentioned OpenAI has “a whole lot of different essential issues to launch” earlier than its upcoming GPT-5.
These options may not come to fruition as corporations usually file trademark purposes for concepts that by no means see the sunshine of day. OpenAI is, nonetheless, anticipated to launch a “materially higher” improve of its ChatGPT mannequin mid-year, as Enterprise Insider beforehand reported.
In an interview with podcaster Lex Fridman, Altman mentioned: “We are going to launch an incredible new mannequin this yr. I do not know what we’ll name it. We’ll launch over within the coming months many various issues, I believe they will be very cool.”
He added: “I believe earlier than we discuss a GPT-5 like mannequin known as that or not known as that or a little bit bit worse or little bit higher than what you’d anticipate from a GPT-5, I believe now we have a whole lot of different essential issues to launch first.”
The corporate would not but supply a digital voice assistant to its customers, though it does have an API that converts speech to textual content known as TTS, in accordance with its web site. It additionally has a general-purpose speech recognition mannequin known as Whisper.
OpenAI additionally filed trademark purposes for its future fashions, together with GPT-6 and GPT-7 in October.
The submitting for GPT-6 consists of simulating conversations, sharing datasets for the aim of machine studying, predictive analytics, and “analyzing algorithms which might be in a position to study to research, classify, and take actions in response to publicity to knowledge.”
The GPT-7 trademark software covers software program utilizing synthetic intelligence for music era, conversion of textual content and knowledge information into software program code, and creating and producing software program code. Each purposes are nonetheless underneath examination.
The Patent Workplace may not grant these emblems, although, as in February, it rejected OpenAI’s makes an attempt to trademark “GPT.” The company mentioned it is a “broadly used acronym” that is “merely descriptive” because it means “generative pre-trained transformers.”
The “voice engineer” trademark software, which is pending a evaluation from an examiner, intends to cowl software program for the next 10 areas:
-
Constructing digital voice assistants
-
Voice and speech recognition, processing voice instructions, and changing between textual content and speech
-
Processing voice instructions, and changing between textual content and speech
-
automated speech and voice recognition and era
-
Producing voice and audio outputs primarily based on pure language prompts, textual content, speech, visible prompts, photos, and/or video
-
Technology of audio and/or voice in response to person prompts
-
Machine learning-based pure language and speech processing
-
Multilingual speech recognition, translation, and transcription
-
Utilizing synthetic intelligence for automated text-to-voice and text-to-audio conversion
-
Improvement of voice service supply
OpenAI did not instantly reply to a request for remark from Enterprise Insider.
