No Text-to-Speech Option Available in Codex Mobile

Codex Feature Requests
,
1

Apparently this has been a limitation for almost 5 months. This is a non-negotiable feature for me. Claude’s text-to-speech is weird, but I am about to cancel my ChatGPT subscription due to this limitation and move to Claude remote control.

OpenAI, please understand that Voice Mode is cool, but that is not a solution for everyone. Some people fundamentally talk and think more asynchronously than that. Sometimes I need to actually think about what I’m going to say next and I can’t handle that assistant constantly interrupting me because I took a 3-second pause. I need to use whisper to write a message, then hear that response spoken aloud in a single go.

Also, reading messages on my screen while I’m driving is not an option… and that’s why I’m switching to Claude.

2

Thanks for sharing this — your use case is very clear, and I think the distinction between live voice and asynchronous voice interaction is especially important.

Being able to speak a request, let Codex work, then simply listen to the response without needing to keep looking at the screen could make the mobile experience much more practical, especially while multitasking.

Interestingly, I was thinking about almost the same problem just before seeing your post. I’m currently exploring a project around improving voice-based human–AI interaction and making it more efficient for real work, particularly when combined with agents and computer interaction.

Your example is a very useful real-world signal. Thanks again for sharing it.