OpenAI has made available APIs for ChatGPT and the company’s AI-powered transcription and translation service Whisper. These APIs will help businesses to integrate ChatGPT and Whisper into their conversation platforms and will be substantially cheaper than using the company’s existing language model.
“Through a series of system-wide optimizations, we’ve achieved 90% cost reduction for ChatGPT since December; we’re now passing through those savings to API users,” the company said in a blog post.
OpenAI launched Whisper in September and ChatGPT in November. Now with this API offering, ChatGPT and Whisper can now be officially integrated into third party software.
“ChatGPT and Whisper models are now available on our API, giving developers access to cutting-edge language (not just chat!) and speech-to-text capabilities,” the company said.
Speech-to-text with Whisper
Whisper enables transcription in multiple languages, as well as translation from those languages to English.
Whisper has garnered massive praise from the developer community but can also be “hard to run”, OpenAI admitted. The company has “now made the large-v2 model available through our API, which gives convenient on-demand access priced at $0.006 / minute (of audio input),” it said.
Whisper’s large-v2 model in the API provides much faster and cost-effective results, OpenAI said.
Whisper accepts files in multiple formats including M4A, MP3, MP4, MPEG, MPGA, WAV and WEBM. It is trained on 680,000 hours of multilingual and multitask supervised data collected from the web.
ChatGPT API based on GPT 3.5 Turbo
The ChatGPT APIs sit on the GPT 3.5 Turbo language model, which is the same “family” as the one used in the latest version of ChatGPT.
Companies using ChatGPT for their operations will be required to pay $0.002 per 1000 tokens (about 750 words). OpenAI will provide businesses with a tool to assess how many tokens would per user query cost and translate this to match their customer query scale.
OpenAI claimed that their pricing is 10 times cheaper than the existing GPT 3.5 models, the version which originally powered ChatGPT.
“ChatGPT API users can expect continuous model improvements and the option to choose dedicated capacity for deeper control over the models,” OpenAI said.
ChatGPT popularity continues to grow
ChatGPT was launched in November and was an instant hit. It registered 100 million active users just two months after its launch. However, due to this enormous demand, the company could not keep up with the pace leading to downtimes.
For the past two months ChatGPT uptime has not met its own expectations “nor that of our users. Our engineering team’s top priority is now stability of production use cases,” OpenAI said.
Several organizations including technology giants like Google, Amazon, Apple, and Meta have created advanced speech recognition systems. If OpenAI can establish itself in the speech-to-text market in a big way, it may prove quite profitable for the Microsoft-backed company.
The text to speech market could be worth $5.4 billion by 2026, up from $2.2 billion in 2021, shows a report by MarketsandMarkets.
AI can give “incredible opportunities” and “economic empowerment to everyone,” Open AI said.
Copyright © 2023 IDG Communications, Inc.
Discussion about this post