Hacker News new | ask | show | jobs
by marc 1206 days ago
Congrats on the launch!

Made something similar which also uses Whisper to support voice memos (talk to GPT) and TTS (hear GPT’s responses)

Not sure it warrants a separate post, so sharing it here.

https://t.me/marcbot

5 comments

Hey that's pretty awesome, can you make the TTS output language configurable?

Also when i send something to the bot i can only send video, (with the iOS app) it makes more sense to send audio. Is this something you can change?

Finally, you should have a privacy policy for a service like this.

Yes, you can configure the TTS voices (and languages).

If you send /voices you can pick from multiple voices.

Right now it only shows voices in English, but you can send the secret `/setvoice <voiceName>` with an Amazon Polly[0] neural voice and it will work as well.

The only downside right now is that I don't auto-identify languages, so if you set it to Dutch, but ask a question in English, you'll get a response in English with a very Dutch accent haha.

P.S. And yes @aero-glide2 is correct that you can toggle between Telegram's audio/video inputs by tapping the camera/microphone icon. – Right now MarcBot only supports audio inputs.

[0] https://docs.aws.amazon.com/polly/latest/dg/voicelist.html

I love this - thanks! Easiest interface into ChatGPT I've found.

/setvoice didn't see to work for me.

I was looking for a British voice and tried

/setvoice Brian

Am I doing something wrong?

Tap on video once, it will change to audio.
or autodetect the output language. Both Whisper and ChatGPT handle multilingual just fine
What do you use for TTS? I'm designing an in-app chatbot and having a hard time figuring out a fast-enough TTS library.

Currently considering either pre-recording all the outputs (ugh) or using the browser's TTS api (ugh^2) :)

Ah, it looks great. It's just 100% English though, which doesn't fit my use case (customers in 6 languages)
Thanks, I saw it yesterday and tried it, it's pretty awesome! Well done!
Does this also use chatgpt's api? Just tried it, it's very useful.
Yep! Same ChatGPT API for text generation. And Whisper AP for speech recognition.

If you speak other languages than English, make sure to give those a try well. It should understand you and respond in the same language.

Can you share the source code?