Feature request: speech-to-text & text-to-speech

[zenflow]

It would be really cool if this app supported voice conversations with ChatGPT. Here's a chrome extension that enhances the official site with that feature: https://github.com/C-Nedelcu/talk-to-chatgpt

[lencx]

Its voice seems to be provided by chrome, so it may not be possible to do it without chrome. I can look into it.

[mefengl]

expose an API that can be called from Shortcuts will be great, so that can be integrated with Siri and other automation

[Ethkuil]

Its voice seems to be provided by chrome, so it may not be possible to do it without chrome. I can look into it.

It's all right to use other APIs. It would be much more appealing for people who want to practice listening and speaking of a foreign language. You know, for example to prepare for TOEFL and IELTS.

[cyhhao]

Great idea, I'm trying to practice my English with ChatGPT.

Maybe I can take some time to help this author with this feature.

[melbarra88]

As an improvement to this feature, I suggest allowing the user to change the voice in the chat window for each message (overriding the default voice configured globally).

I usually use different languages when using Chat GPT, and it would be great if I could change the synthesis voice...

It would be even better if the language of the message was automatically detected and a voice corresponding to that language was automatically selected for the speech synthesis.

[code-whale]

Can users use their own API key for speech synthesis in future versions, such as using Azure?

[zenflow]

Can users use their own API key for speech synthesis in future versions, such as using Azure?

Ideally it wouldn't use a paid service for this and you wouldn't need any API key.. Speech synthesis and speech recognition can be done on local machine.

[code-whale]

@zenflow Thank you for your reply, but please forgive me, for only representing my personal opinion that the current speech synthesis is not perfect and sounds too 'mechanical' in some pronunciations. Therefore, I hope to use my own API for speech synthesis. I think Azure's speech synthesis is very suitable for people like me who want to practice English speaking and listening skills:)

[zenflow]

that the current speech synthesis is not perfect and sounds too 'mechanical' in some pronunciations

I had not considered a reason like this for using a paid service for speech synthesis! It makes sense!

That said, I hope that an API key is not required for those of us who are fine with the mechanical speech synthesis our local machines can do by themselves.

[Gourdbaby]

have this feature solved? I am also a person who want to practice English speaking skills. If this app can support the feature, I would very appreciate it.

[shawn-ann]

I also want this feature