OpenAI is looking for a communications engineer for the team behind GPT-4o, the chatting ChatGPT. According to the company’s employment platform, the salary range for this vacancy is between $310,000 to $465,000 per year.
The job description highlights that it is the first role of this type that OpenAI wants to hire. “We are looking for an engineer with real-time communications (RTC) experience to join OpenAI as our first dedicated RTC specialist”:
The startup that has investment from Microsoft announced its new GPT-4o model, capable of interacting with users through voice in real time. For example, this AI is capable of translating conversations between two people who speak different languages.
For this release, OpenAI is looking for a real-time communications engineer, in order to polish GPT-4o.
In the description they mention that the job is in San Francisco, California, however, in another section that describes the position, OpenAI asks the candidate: “Will you need sponsorship now or in the future to obtain employment visa status in the United States?” Joined?”. For this reason, it is not clear if it is a face-to-face job or if it can be remote at first.
The technology company also details the responsibilities of this position:
- Innovate techniques to integrate artificial intelligence and real-time communication, ensuring minimum latency.
- Establish new standards for the transmission of audio and video inputs and outputs between users and models.
- Lead cross-functional efforts and alignment between infrastructure, product and research teams.
The company behind the popular ChatGPT chatbot suggests that candidates for this communications engineer position have the following characteristics:
- Have extensive knowledge of WebRTC, audio and video codecs, and low latency streaming systems.
- Have experience building production software systems at scale.
- Experience with AI models and big language is a plus, but not required.
Apart from the salary range of $310,000 to $465,000 annually, the selected person has access to:
- Medical, dental and vision insurance for the employee and his or her family.
- Mental health support.
- Paid parental leave (20 weeks) and family planning support.
The artificial intelligence model for which OpenAI is requesting a communications engineer is characterized by audio input and output, that is, the AI can understand and respond to you with its own voice.
GPT-4o has the ability to produce content and understand commands in voice, text or image formats.which significantly expands the options for interaction between humans and artificial intelligence.
The company shared a video in which the new model was tested. It consisted of a conversation between two people, one who spoke Italian and the other English. OpenAI showed that GPT-4o had the ability to translate both languages in real time.
This artificial intelligence will be available for free to all ChatGPT users in the coming months.
ChatGPT’s voice has also been a hot topic because the renowned Marvel actress, Scarlett Johansson, expressed her disagreement due to the similarity with her tone of voice.
ChatGPT has five different voices: Breeze, Cove, Ember, Juniper and Sky. The latter is the one Johansson complained about.
“My friends, family and the general public realized how much the new system called Sky was like me. Before we could connect (with Sam Altman, the system was already on the market), Johansson published in a statement.
In early 2023, OpenAI worked with casting directors and producers in this industry to establish specific criteria for their AI voices.
To address concerns regarding Sky’s voice and safeguard the integrity of voice authors hired by the startup, OpenAI decided to temporarily suspend this voice on ChatGPT.
