OpenAI is launching new voice intelligence features in its Application Programming Interface (API).

OpenAI said Thursday that its application programming interface (API) will now include a number of new voice intelligence features designed to help developers create apps that can speak, record and translate conversations with users.

The company’s new GPT‑Realtime‑2 is another voice model, designed to create realistic voice simulations that can converse with users. However, unlike its predecessor (GPT-Realtime-1.5), this version is built using GPT-5 class logic that OpenAI says was created to handle more complex requests from users.

The company is also launching the GPT-Realtime-Translate service which, apparently, is designed to provide real-time translation services that “keep up” with the user during the conversation. The feature includes more than 70 input languages ​​(i.e. the languages ​​it can understand) and 13 output languages ​​(the languages ​​it transmits to the speaker).

Finally, the company also launched a new transcription capability, GPT-Realtime-Whisper, which gives users live speech-to-text capabilities that are captured while interactions are taking place.

“The models we are launching together take real-time audio from simple call and answer to voice interfaces that can actually work: listening, thinking, translating, transcribing, and taking action as the conversation unfolds,” the company said.

Who will these updates be useful to? Companies that want to expand their customer service capabilities are an obvious target. However, OpenAI also notes that its new features will help in a wide range of fields, including education, media, events, and creator platforms, among others.

As useful as these tools are from an organization’s perspective, they can also be abused. The company said it has built guardrails to prevent its new features from being misused to create spam, fraud or other forms of online abuse. Certain triggers are built into the system so that “conversations can be stopped if they are detected as violating our harmful content guidelines,” OpenAI said.

All new audio models are included in OpenAI’s Realtime API. Translate and Whisper are billed by the minute, while GPT-Realtime-2 is billed by token consumption.