What began as a simple experiment has evolved into a fully operational AI voice assistant, designed to assist home cooks with hands-free recipe following. The tool is available on a well-known recipe website.

The project was spearheaded by Hearst’s senior director of AI initiatives, Alexandria Redmon, who envisioned a personal chef assistant that could be interacted with through voice commands, eliminating the need to touch a device with messy hands. Using OpenAI’s GPT large language model,Redmon developed the prototype in a matter of days. The assistant is designed to tackle common kitchen challenges, such as managing multiple timers, suggesting ingredient substitutions, adn providing hands-free recipe navigation.

Redmon collaborated with editorial director Joanna Saltz and director of product Ashley Szwec.Together, they dedicated five months to developing the “Cooking Coach” voice assistant, which has now been launched on the website. The AI is trained using a database of over 30,000 recipes from the site’s ten-year history.

The “Cooking Coach” is the frist AI product for the site. A pop-up invitation encourages users to enable thier microphone to access the assistant. Once activated, the voice tool assists users in selecting and following recipes. For example, a user might say, “I’m looking for chicken tacos that take less than an hour.” The assistant can also recommend ingredient swaps, adjust serving sizes, set multiple timers, and answer questions about cooking techniques and terminology, all in multiple languages.Saltz mentioned that perfecting the voice assistant took some time, as the initial voice had a “Scottish brogue.”

Publishers are increasingly developing generative AI products, including chatbots and search assistants, to encourage users to spend more time on their websites and to promote their own content. This push comes as AI products and search engines are increasingly capturing users’ attention by providing swift, simplified answers that bypass publishers altogether.

According to Josh Jaffe, AI and media consultant and former president of media at the publisher Ingenio, “User expectations are shifting from reading content to relying on AI tools that help them complete tasks step by step. Publishers need to meet that shift by building interactive, utility-driven experiences. Tools like this are one way to do that.”

user expectations are shifting from reading content to relying on AI tools that help them complete tasks step by step.

As Google incorporates more generative AI summaries into its search engine, lifestyle publishers face the challenge of maintaining stable traffic amid declining search referral traffic. Google’s AI-generated summaries can provide complete recipes to users, posing a particular risk to recipe websites.Saltz noted that they have seen “a little bit of an impact” on their traffic due to these Google AI features, but remains optimistic.

Saltz believes that “Coming out with innovations that are different and fascinating and fun and make cooking easier… is going to ensure that we’re relevant for a long time.”

Voice assistants are becoming more smart with the help of generative AI technology. Amazon is reportedly seeking AI licensing partnerships with news publishers to enhance its voice assistant, Alexa. The New York Times’ recent AI licensing agreement with Amazon suggests that its content may power the new Alexa+.

Redmon chose not to comment on whether they were in discussions with Amazon regarding making their content and AI voice initiatives available through Alexa.However, she acknowledged the potential value of device-specific integrations.

“Rather of limiting myself to just folks that have Alexa… I at least wanted to be able to prove out what could be done,” said Redmon. “And so making it a mobile-optimized web app was a way to reach basically everybody in this space at launch.”

the team is monitoring engagement metrics to assess the success of the Cooking Coach. Ashley Szwec added that the team can also analyse chat logs with the Cooking Coach to understand how users are interacting with recipe discovery. The team intends to expand features that prove popular with users in the future.

Jaffe stated, “It’s not groundbreaking tech, but it’s a smart start. A next-level version would integrate predictive personalization and contextual tips based on user behaviour, turning it into a true personal sous chef. That’s the kind of AI-native product publishers need to be building now if they want to stay competitive.”

The team is also exploring the use of AI technology to improve the search function on their site. this includes using natural language search to enable users to find recipes that exclude specific ingredients or use terms like “vegetarian,” even if those terms are not included in the recipe title.They are also considering developing photo-based search, where users can upload a picture of items in their pantry and ask the cooking assistant for recipe suggestions.

Ronak Patel, gm of the lifestyle group at Hearst Magazines, said, “This is just the beginning.”

Jaffe believes that users will only prefer a publisher’s AI experience over a generic ChatGPT result if the UX, tone and content stand out. “That is where publishers still have an edge,” he added.