[ad_1]
OpenAI has launched ChatGPT and Whisper speech-transcription APIs for developers. Any interested individual or organization can now leverage these groundbreaking AI technologies to improve their products or develop new products from scratch. Snapchat’s newly announced My AI is one of the early products powered by the ChatGPT API.
According to OpenAI, its ChatGPT API will use the “gpt-3.5-turbo” AI model, the same used by the chatbot itself. This allows developers to integrate an unchanged version of the tool into their products while also giving them the flexibility to tweak a few things. The company says that developers can use it to build conversational AI chatbots as well as make their “non-chat” platforms more intelligent.
The ChatGPT API will cost developers $0.002 per 1,000 tokens, which equates to about 750 words. This is the default plan and gives developers a shared computing infrastructure for processing requests. However, OpenAI is also offering “dedicated instances for users who want deeper control over the specific model version and system performance”. They can reserve an allocation of the infrastructure by paying for a time period regardless of how many requests they send. This is the ideal solution for those who process more than 450 million tokens per day, the company states.
Along with Snap, a few other brands have already launched products powered by the ChatGPT API. Online learning platform Quizlet is using the tech in its adaptive AI tutor Q-Chat. Likewise, grocery delivery chain Instacart and Shopify’s consumer app Shop also leverage OpenAI’s technology to offer new experiences to their consumers. OpenAI has already launched a consumer-facing premium version of ChatGPT. Launched last month, the $20-per-month service is currently available to users in the US. ChatGPT Plus brings faster response times and priority access.
OpenAI also launched Whisper API
Along with ChatGPT, OpenAI is also giving developers access to its Whisper speech-to-text model with a new API. The company launched an open-source model of it in September last year. It has now updated it to a large-v2 model and made it available to developers through its API. This speech recognition system can transcribe and translate multiple languages that you can feed in various file formats.
“We released a model, but that actually was not enough to cause the whole developer ecosystem to build around it,” OpenAI co-founder Greg Brockman told TechCrunch. “The Whisper API is the same large model that you can get open source, but we’ve optimized it to the extreme. It’s much, much faster and extremely convenient”. Developers can get on-demand access to this speech-transcription API at $0.006 per minute. AI-powered language learning app Speak is already using Whisper API, OpenAI revealed.
Last but not least, OpenAI has updated its API terms of service to give customers peace of mind when it comes to data privacy. The company says that it will not use any data submitted through the API for service improvements. Organizations can manually opt-in and provide their data set for model training and other uses, though. It has also announced a default 30-day data retention policy for API users. Once again, developers have “options for stricter retention” depending on their needs.
“We believe that AI can provide incredible opportunities and economic empowerment to everyone, and the best way to achieve that is to allow everyone to build with it. We hope that the changes we announced today will lead to numerous applications that everyone can benefit from. Start building next-generation apps powered by ChatGPT & Whisper,” OpenAI said in a blog post.
[ad_2]
Source link