ChatGPT and Whisper API update offers faster and cheaper language and speech-to-text capabilities

ChatGPT and Whisper API update offers faster and cheaper language and speech-to-text capabilities

Developers now have access to cutting-edge language and speech-to-text capabilities with the availability of ChatGPT and Whisper models on OpenAI's API. The system-wide optimizations have led to a 90% reduction in cost for ChatGPT since December, and the savings will be passed through to API users. ChatGPT model family released today (gpt-3.5-turbo) is priced at $0.002 per 1k tokens (750 words), which is 10x cheaper than existing GPT-3.5 models.

The use of the open-source Whisper large-v2 model in the API provides developers with faster and cost-effective results. In addition, ChatGPT API users can expect continuous model improvements and the choice of dedicated capacity for more control over the models.

The API has also refined its terms of service in response to developer feedback. Dedicated instances are now available for users who want more control over the specific model version and system performance. Developers can pay for an allocation of compute infrastructure that is reserved for serving their requests. The instances provide full control over the load, the option to enable features such as longer context limits, and the ability to pin the model snapshot. For developers running beyond ~450M tokens per day, dedicated instances can make economic sense and directly optimize workload against hardware performance.

Whisper, the speech-to-text model that was open-sourced in September 2022, can be challenging to run, but the availability of the large-v2 model through the API gives convenient on-demand access priced at $0.006/minute. The API is available through transcriptions or translations endpoints and accepts various formats, including m4a, mp3, mp4, mpeg, mpga, wav, and webm.

Feedback from API customers has been essential to understand how to serve them better, leading to changes such as the default 30-day data retention policy for API users, improving developer documentation, and simplifying the terms of service and usage policies. The API no longer uses data submitted through the API for service improvements, including model training, unless the organization opts in. The input and output of the models are owned by users, as stated in the terms around data ownership.

The engineering team has made stability of production use cases a top priority after the uptime of the API did not meet users' expectations over the past two months. Ensuring AI benefits all of humanity requires being a reliable service provider, and the team is working hard to improve uptime over the upcoming months. The changes announced today are expected to lead to numerous applications that everyone can benefit from, providing incredible opportunities and economic empowerment to everyone. Developers can now start building next-generation apps powered by ChatGPT & Whisper.

Early users of ChatGPT and Whisper APIs:

  • My AI for Snapchat+
    The experimental feature is running on ChatGPT API. My AI offers a friendly, customizable chatbot at their fingertips that offers recommendations, and can even write a haiku for friends in seconds.

  • Quizlet
    Quizlet has worked with OpenAI for the last three years, leveraging GPT-3 across multiple use cases, including vocabulary learning and practice tests. With the launch of ChatGPT API, Quizlet is introducing Q-Chat, a fully-adaptive AI tutor that engages students with adaptive questions based on relevant study materials delivered through a fun chat experience.

  • Instacart
    Is augmenting the Instacart app to enable customers to ask about food and get inspirational, shoppable answers. This uses ChatGPT alongside Instacart’s own AI and product data from their 75,000+ retail partner store locations to help customers discover ideas for open-ended shopping goals, such as “How do I make great fish tacos?” or “What’s a healthy lunch for my kids?” Instacart plans to launch “Ask Instacart” later this year.

  • Shop
    ChatGPT API is used to power Shopify’s new shopping assistant. When shoppers search for products, the shopping assistant makes personalized recommendations based on their requests. Shop’s new AI-powered shopping assistant will streamline in-app shopping by scanning millions of products to quickly find what buyers are looking for—or help them discover something new.

  • Speak
    They’re the fastest-growing English app in South Korea, and are already using the Whisper API to power a new AI speaking companion product, and rapidly bring it to the rest of the globe. Whisper’s human-level accuracy for language learners of every level unlocks true open-ended conversational practice and highly accurate feedback.

More news
Tags: