wanted to share some frustrations I’ve been experiencing with the ChatGPT app. For over a week now, the read-aloud feature hasn’t been working, which has been a big disappointment. I chose this app largely because of that feature, as it’s something I use frequently.
On top of this, I’m noticing that many features come with strict usage limits. Given that ChatGPT is priced as a premium service, I find these restrictions surprising, especially since other AI platforms offer similar features without these limits.
I’m curious to know if others feel the same way about these limitations or if anyone has heard about potential updates to address them. For the investment we’re making, I expected a more unrestricted experience. What’s everyone’s take on this?
At first, I thought the same: paying for an app with issues?!
But I’ve tried all the AI available to me (except humanoids—I’m too poor to afford that), and right now, ChatGPT is by far the best on the market. AI is taking off, and no system is perfect yet, neither ChatGPT nor any other.
Despite its flaws, it’s still great. Once it can have more memory and user personalization, it’ll be unbeatable by a long shot.
I’m almost updating the app daily, which shows just how hard they’re working to improve it day by day. In my case, whenever I’ve reported an issue, it’s been solved and improved in a short time. Maybe that will help you.
With the full context that AI is in its early stages and isn’t perfect yet anywhere or with any company, that there isn’t a better app right now, that developers are tirelessly working to improve it daily, and that the cost is likely higher than what they make from servers and employees working non-stop… I reached the conclusion that it’s worth paying for ChatGPT.
I use it for thousands of things. If you’re creative, it offers an enormous range of possibilities.
I encourage you to explore what ChatGPT can do for you. And as for the reading limit… it’ll surely improve soon.
I’m reaching out to share some technical issues I’ve recently encountered with the Read Aloud and Voice Mode features in ChatGPT. Since the recent launch of the search feature, I’ve noticed some consistent errors with the voice capabilities that weren’t present before.
Here’s a summary of what I’ve experienced:
Read Aloud Not Working: I often receive a “Resource Not Found Error” when trying to use the Read Aloud option. Occasionally, I’ve seen other error codes, possibly including 404, though I’m not entirely sure of all the specifics. Regardless, the feature is breaking frequently and failing to function as expected.
Voice Mode Issues: When I use the Voice Mode for conversation, the system often breaks or stops unexpectedly right after it begins to speak. This issue has only started occurring recently and disrupts the experience significantly. (GPUs are overloaded? )
Voice Recording Errors: When recording messages, I sometimes experience errors that lead to lost recordings. For instance, I recorded a message lasting about five minutes, but afterward, the recording was gone with no data saved. This has happened intermittently and wasn’t an issue before (This one is kind of irritating).
It’s worth noting that these problems started appearing around the same time as the search feature release, so I wonder if there’s a connection to server capacity or other resource limitations.
If anyone else has reported similar issues or if there’s a solution in the works, I’d appreciate any updates.
Error: CoreMediaErrorDomain error -16830 with ChatGPT Read Aloud Feature”
Hello everyone,
I’m encountering an issue with the ChatGPT app’s “Read Aloud” function on my iOS device, and I’d appreciate any insights or solutions.
Issue Description: When I try to use the “Read Aloud” feature to listen to a response, I get an error message that says:
Error: ‘The operation couldn’t be completed. (CoreMediaErrorDomain error -16830)
Troubleshooting Steps I’ve Tried:
Restarted the app.
Checked that my iOS device and the ChatGPT app are both updated to the latest versions.
Despite these steps, the issue persists, and I can’t use the Read Aloud feature. I’m wondering if anyone else has encountered this error or knows what might be causing it.
I’ve identified an issue where the text-to-speech feature struggles with longer texts. For instance, when you input a short phrase like ‘Hi’ and select ‘Read Aloud,’ it functions correctly. However, when the text is longer, the feature fails to work as expected. This indicates a possible limitation with handling extended text inputs
My problem is not resolved after doing complaints in each available platform now it’s been four or five days my work stopped because of features are not working accordingly it’s 2 days when I submit my complaint and after that they ask me to provide the photo of my problem and I will provide the photo and just half an hour and its been 2 days since then I have no progress in
And the feature Read aloud and I also tried by reinstalled the app but nothing changed and I am not get getting any response from the customer service team that’s why I have to take this problem to the community and this not a first time when I have to take my problems to the community because the customer service does not response to my problems I does not expect that a platform like GPT should ignore the problems of his customers
This issue has already been brought to the attention of the OpenAI staff, meaning they are aware of the problem and we are awaiting further action.
I understand it must be frustrating to have the Read Aloud feature not working properly, but given the nature of the issue, it might take some time to resolve.
It’s been 2 days my read aloud it’s not working when I contact the chat about responsible for customer responsive I received and mail saying that provide the photo of the issue and it’s been 5 to 6 hours I provided the photo but no results is coming because the feature is not working properly my work has been slow down if I write something from the voice mode appeared in the chat but when I continue the story from typing or using another method it does not take the part created by using the voice mode and same goes for the voice modify write something by typing or the another method the voice phone cannot take it into the response and I feel another problem also the chat GPT response memory is too small if I am making to stories at one time the Jas GPT only take the part of main points and sometime even mix it up we understand that the cheque GPT cannot providers unlimited memory but they can connect the charge of eating memory to our own storage so that we can easily do or work without wearing about the memory
I was checking everything prior to recording a training video when I discovered the same message on my Android mobile phone.
It IS working on the web version on my PC and on the iOS app on my iPad
On my Android phone, I managed to select the text and use ReadAloud (with Libby) in Edge to listen to the corrected transcript even though the speaker icon didn’t work. There was no error message unlike on the Android App which said, “Sorry, something went wrong.”
I’m having the same exact issue my problem is I’m using voice on my videos because it’s the best voice out there and it’s free anyway I’ve been waiting patiently now for about a week for a fix or a patch same exact issue and my Android Motorola phone is just not working but in any browser extender I use it does
OpenAI offers an API service for receiving spoken audio from transcripts. It can run at high rate on segments in parallel.
That you are using a middleman AI, to which you simply say “repeat back to me” at OpenAI’s computational expense and no actual AI need, so that you only would be able to then send text to the same type of TTS service as is on API, is exactly the kind of thing that may make OpenAI reconsider what is offered in ChatGPT.
I am encountering a persistent issue with the “Read Aloud” feature in the ChatGPT app on my iPhone since updating to the iOS 18.2 beta. The error prevents the “Read Aloud” function from starting, and it displays the following message:
“The operation couldn’t be completed. (CoreMediaErrorDomain error -16830)”
Occasionally, the feature works once but fails again on subsequent attempts. I’ve detailed my device setup and troubleshooting steps below.
Device and App Information:
• Device Model: iPhone 16 Pro Max
• iOS Version: iOS 18.2 Developer beta (22C5109p)
• ChatGPT App Version: 1.2024.296 (11526659645)
• Account Type: ChatGPT Plus subscriber
Troubleshooting Steps Taken:
1. Checked Microphone Permissions: Confirmed that microphone access is enabled for the ChatGPT app.
2. Restarted Device: Rebooted the iPhone multiple times.
3. Reinstalled App: Deleted and reinstalled the ChatGPT app from the App Store.
4. Network Testing: Attempted on both Wi-Fi and cellular networks to rule out connectivity issues.
5. Checked OpenAI’s Status Page: Verified that there were no reported issues on OpenAI’s status page.
Additional Information:
This issue began immediately after updating to iOS 18.2 beta, which includes new features powered by Apple Intelligence. I suspect there may be a compatibility conflict between the iOS 18.2 beta and the ChatGPT app’s media playback functionalities, specifically around CoreMedia frameworks used for audio playback.
Any guidance on how to resolve this issue, or if others in the community are experiencing similar problems with iOS 18.2, would be greatly appreciated. If further diagnostic information is required, please let me know!
I can also confirm this error. I’ve seen it for the first time yesterday and today. I think it has to do with character generation. Possibly, any generations of around 6k or more will fail to produce a successful TTS (Speak Aloud) generation. I’ve never had this issue before. Does this have to do with updates to ChatGPT 4o? Maybe ChatGPT 4o had a smaller character limit and now it’s larger but the TTS function wasn’t updated with it?
I think it’s happening to everyone because I have 2 devices that have chatgbt and 2 different accounts and it’s the same problem with the “read aloud” feature
30/10/2024 - MOTO G52 - Same issue since 2 days ago! I used to use this feature while I was driving… it’s a pity that it isn’t working anymore after an update. Also, the voice has changed. Everybody sounds like a Spanish accent not neutral.
)
Restarted the app.
