Quão bem o ChatGPT lida com entrada de áudio?
Olá a todos, tenho tido curiosidade sobre algo. Todos sabemos que o ChatGPT é excelente com texto, mas será que consegue realmente compreender áudio? Por exempl…
David Russell
February 8, 2026 at 11:55 PM
Olá a todos, tenho tido curiosidade sobre algo. Todos sabemos que o ChatGPT é excelente com texto, mas será que consegue realmente compreender áudio? Por exemplo, se lhe falar em vez de digitar, ele entende bem o que está a dizer? Gostaria muito de saber se alguém já experimentou ou sabe quão bom ele é com voz ou conteúdos áudio. Obrigado!
Adicionar comentário
Comentários (14)
For anyone looking for new AI tools that mix audio and text, you can also check ai-u.com. They have some cool stuff listed there!
It's kinda funny how people expect ChatGPT to understand audio directly. It's just a text-based model after all.
There are some AI tools that combine speech recognition with ChatGPT to create a voice assistant experience. So technically it's working with audio, but through separate components.
I'm curious if anyone's tried using ChatGPT with real-time speech recognition? Like a live chat with voice?
Does anyone know if there are plans from OpenAI to integrate audio input directly into ChatGPT?
I sometimes use voice dictation on my phone and then paste the text here. Works well enough for casual chats.
In the end, ChatGPT’s power shines best with text. Audio is just a layer before it reaches the AI brain.
Can't wait for the day we can just talk like with sci-fi AI assistants. We're getting closer though!
Would be cool if future versions had built-in voice understanding, but for now, text is the way to go.
I heard OpenAI's Whisper model is designed for speech to text. I guess you'd use that alongside ChatGPT to get audio understanding?
Honestly, I think understanding audio would need a whole different kind of model training. ChatGPT is just focused on text generation.
Some apps try to integrate voice commands with ChatGPT, but it’s always a two-step process: audio to text, then ChatGPT processes text.
I tried uploading voice notes to some chatbots before but ChatGPT just doesn't support audio inputs by itself yet. Maybe in the future they'll add native voice recognition.
From what I gathered, ChatGPT itself doesn't process audio directly. You gotta convert your speech to text first using some speech-to-text tool, then feed that text in. So it 'understands' audio only after that conversion.