OpenAI has announced that it is giving its ChatGPT chatbot the ability to communicate with users using voice and audio, putting the company in direct competition with tech giants like Google, Apple, and Amazon in the race to create smarter voice assistants. ChatGPT can now respond to user queries in a more conversational tone, thanks to its ability to assume different “personas.” OpenAI sees the addition of voice functionality as a way to encourage more people to interact with and use ChatGPT.
Peter Deng, OpenAI’s VP of consumer products, stated that the challenge lies in simplifying the technology so that it appeals to the broader population. OpenAI’s latest announcement highlights Amazon’s lag in launching new AI tools for the general public compared to other tech giants. On the same day as OpenAI’s announcement, Amazon revealed a deal to invest up to $4 billion in AI startup Anthropic, the largest investment in this space since Microsoft’s landmark investment in OpenAI earlier this year.
Last week, several companies, including Google, Amazon, and OpenAI, launched various AI products, indicating the frenzied pace of competition in the market. Introducing voice assistants into existing speakers is a key strategy being explored by companies to encourage user adoption and monetization. Amazon recently added a chatbot “conversation” feature to its Alexa speakers, and Google is considering integrating its Bard chatbot technology into its voice assistants.
Previously, users could ask ChatGPT questions verbally on the mobile app, but the chatbot would respond with text only. OpenAI has now expanded the capabilities of ChatGPT, allowing users to upload images as part of their questions. This development moves ChatGPT closer to becoming a fully “multimodal” model that can “see” and “hear” the world, responding with voice and images, in addition to text prompts. Industry experts believe that creating capable multimodal models is the next frontier in AI research and development, with companies fiercely racing to be at the forefront.
Voice assistants have been widely adopted in devices like cars, smartphones, TVs, and home speakers, primarily for simple tasks such as requesting weather reports or turning off lights. However, the introduction of large language models like chatbots opens up the possibility of voice assistants engaging in longer, more natural conversations and answering complex questions.
Amazon has been criticized for its slow response to competition in the generative AI field, but the recent Anthropic deal will give the company access to researchers and technology from the startup. Anthropic, founded by former OpenAI employees, has previously received investments from Google.
OpenAI triggered the chatbot boom when it launched ChatGPT to the public in November. Since then, tech giants have been rushing to develop their own chatbot solutions. Microsoft partnered with OpenAI to leverage its technology, while Google released its Bard chatbot.
AI researchers have warned that users are likely to anthropomorphize chatbots, as their responses often resemble human-like answers. This could lead to the false perception that the chatbot is highly intelligent and capable. Chatbots frequently generate false information and present it as real, a phenomenon referred to as “hallucinating” by AI researchers.
The new personas available for ChatGPT, named Sky, Ember, Breeze, Juniper, and Cove, each have different tones and accents. These voice personas are not intended to resemble any specific person. OpenAI demonstrated how the chatbot can understand meandering and open-ended voice queries, making conversations more fluid and natural.
OpenAI has extensively tested the voice and image features of ChatGPT to ensure appropriate responses to sensitive topics. For example, if users inquire about mental health, the bot will suggest consulting a professional. However, OpenAI acknowledges that there is more work to be done in this area.
Denial of responsibility! Vigour Times is an automatic aggregator of Global media. In each content, the hyperlink to the primary source is specified. All trademarks belong to their rightful owners, and all materials to their authors. For any complaint, please reach us at – [email protected]. We will take necessary action within 24 hours.
