I wanted to know if the regular chat gpt version would get any updates. I've seen chatgpt 4o mimicking humans very well, like with all the pauses and tones. Is the regular chat gpt able to do that too?
#chat gpt 4o
dreamy notch
torpid novaBOT
Hi @dreamy notch. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
The regular ChatGPT and GPT-4o are different versions of OpenAI's language models. GPT-4o is the latest flagship model that can reason across audio, vision, and text in real time. It's designed to process all inputs and outputs through the same neural network, which allows it to handle more complex tasks, such as understanding tone and background noises, which the regular ChatGPT cannot do directly source (https://help.openai.com/en/articles/8400625-voice-chat-faq#h_686ac13528).
However, the regular ChatGPT has been receiving updates to improve its performance and user experience. For instance, OpenAI has updated the performance of the ChatGPT model on their free plan to serve more users and based on user feedback, they are now defaulting Plus users to a faster version of ChatGPT, formerly known as “Turbo” source (https://help.openai.com/en/articles/6825453-chatgpt-release-notes#h_71fed9b7cb).
As for the ability to mimic human-like pauses and tones, the regular ChatGPT writes text and doesn't have direct control or knowledge that the output language is being spoken, unless you tell it. It doesn't know that you didn't type your input. It only is informed if you are using an app, and then to write shorter responses source (https://community.openai.com/t/729163).
In contrast, GPT-4o is designed to handle more complex tasks, such as understanding tone and background noises, which the regular ChatGPT cannot do directly. However, it's important to note that as of the information available, audio support for GPT-4o is planned for the future but not available yet source (https://community.openai.com/t/744455).
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
dreamy notch
----------------------
The regular ChatGPT and GPT-4o are different versions of ...
No no, I think we can use voice conversations with the normal version, at least in mobile, but not as good as the demonstration showed by the Open Al team
frigid escarp
No no, I think we can use voice conversations with the normal version, at least ...
voice has been available for a very long time now. But its just a STT > chat gpt > TTS. Now its just chat gpt
dreamy notch
Wym sst, tts @frigid escarp
half river
Wym sst, tts <@1194683757429522455>
SST = Speech to Text
TTS = Text to Speech
So just to clarify:
The features shown in the livestream and the demo-videos are not available right now.
You can select 4o text model. But most features that were presented will be rolled out in a few weeks or months.
Can i get the emotional voice?
Not yet.
Can i use the 4o text model?
Yes.
Can i use the video-recognition as shown in the livestream?
Not yet. Feature will be rolled out to some users.