I wanted to share with you a program that we’ve made that for Discord takes everything that is said in a voice channel and automatically transcribe it in the to a text channel using the Google speech-to-text API. Basically it is a Discord Speech to Text Bot
See how it works in this video:
As you can see in the video google speech to text is quite accurate even for funny definitely non native English speaker accent. I tested it with some recording of native speakers and it was quite accurate.
There are few use cases that I think that you really help some of you:
Transcribe busy discord voice channels. So you can know what you have missed when you were not listening. We can configure the bot to create mp3 file as well, so you will have a recording of what was said.
It is a lot faster to speak than to type. If time is critical for you / your team, and you are using discord, you can sometimes just prefer to talk to save time.
Use the bot as your own dictating machine. Just say what’s on your mind, and it will give it to you in text, really cool! Can be a huge time saver for bloggers.
If you have podcast, you can use it to convert the discussion to text in PDF format, and give it to your listeners. A great way to attract more following!
Maybe there are more things that can think of, I will happy to know how you plan to use it. Send me a message and let me know :)
I actually made this for myself, because I didn’t find other bots that does this for discord, at least not with all the features that we added:
Easy report of google API billing, so you know that you pay within your budget
If there are other features that you need, let us know, and we will consider adding them.
I think that I should let you know that there are some costs involved:
Google Speech to text API costs $0.006 per 15 seconds (it’s about $1 for 40 minutes so it’s really really good not very expensive). You also will need to create a google developer account and get the API credentials using their service account, and you can do all this privately using a VPN, since there are affordable services you can find for this, from the takethecoupon site online.
You will need a small VPS server that can host the program. If you have one that you can use, great. I currently use the SSD2G from virmach which is good enough for $10/month. Maybe it can work with a cheaper server, I honestly didn’t check. Probably we can convert it to be serverless and use AWS Lambda free tier if there is a demand.
Now, at the moment I don’t plan to make a full service out of this program. To be honest I am not sure it worth my time, as I believe in the (near?) future, social media platform will probably have this feature builtin. Like I said, I made it for myself, but I would be happy to share it with. If you know programming, you probably can do it yourself, but if you don’t, or if you are a developer but too busy to bother, we sell the compiled code for $197, and for extra $50 we will do the installation for you including the ftp so you can download the mp3 recording.
If you have friends that use discord and are looking for a solution like that this, please share it with them, I am sure they will thank you for it as it is a huge time saver!
To purchase the Discord Speech-To-Text Bot use the paypal button below: