New IF interpreter with audio input and output for android


#1

Hi everyone,

I just released an android app called audio Interactive Fiction, which support around 700 games playable in a dynamic way of listen talk flow.
You can check it out here play.google.com/store/apps/deta … os.audioif

If you have any question or suggestion let me know!

I will paste here the app store description also:

Play your favorite interactive fiction games in a listen-talk way.
At each step the text adventure is read for you, and then waits for your voice command to play the game.

2 modes:

-Text mode: regular interactive fiction gameplay with reading and writing
-Audio mode: dynamic listen-talk gameplay

I tried to make the app easy to use for blind people.
The “start audio mode” button is really big in the middle of the screen, so after you open the app, if you press somewhere in the middle of your screen the audio mode should start.
On the audio mode, the app scans for the supported game files on your device, and speaks out loud the name of each game file with a number identifier. After the enumeration of the games, there will be a beep, then you should say the game number, it will load that game.
When it’s time for the player to speak, the app makes a beep sound. You can interrupt the story speech by clicking anywhere in the middle of screen, click anywhere on the screen also to start the speech recognition in case it is stopped.
Sometimes the speech recognition fails to understand what is said, and nothing happens, in that case you have to press in google speech button exactly the middle or dismiss the popup clicking twice somewhere little above the android native back button. (Unfortunately I couldn’t solve this problem yet)

If there is any problem or recommendation please let me know.

Text to speech configuration: It gets the default configuration from your phone, so if you want to change speed of speech or language you can do it on android settings, language and input, text to speech output.

How to easily find games:
If you click on the top button of the app “Browse supported games”, it will bring you to the website ifarchive.org/indexes/if-archiv … zcode.html where the game files are directly available for download. Once the game file is downloaded, you just need to go back to the app and start audio or text mode, and the file scanner is going to find your downloaded games.

Game lists:
ifarchive.org/indexes/if-archiv … zcode.html

ifdb.tads.org/search?sortby=rcu … AZ-Machine

How to play quick guide:
microheaven.com/ifguide/step3.html

Supported formats: Z-Machine games
-.z[1-8]
-.zblorb
-.zlb


(Ruber Eaglenest) #2

Great! It works pretty well!

About suggestions:

  1. I would add a button to introduce text with the traditional virtual keyboard, because in some games, there are words, like passwords and magic that the voice to text analysis just can’t handle. Think of xyzzy, BUYON, FROZT, etc.

  2. Add options to control how the voice to text analysis appears. Right now is quite frenetic to play in groups. There’s not time to think. This is not a real problem because the T2Voice deactivates by itself after a while, or you can stop it pressing the screen. So I think a variety of options to activate it would be ideal.

Good work!


#3

Hi Ruber,

I am very happy to hear that it works good! Thanks!

Also thanks for the suggestions, the suggestion number 1 I haven’t thought about it before, it is indeed important to be able to type in passwords, etc. I will think of a way to add a button to add text from keyboard input without taking too much space of the screen. For now what I would suggest to work around this problem is to use the save command at the moment you need to type the password, open the text mode of the app, restore the game and type the password on the text mode. After the complex text part is done, you can save again, and go back to audio mode.

About suggestion number 2: I am curious, what do you mean play in groups? You play with multiple people talking to the phone like some kind of coop mode? It’s a cool idea! What kind of options would you suggest for improving the gameplay? Should I add some settings to add a time interval between the T2Voice and the voice recognition?

Thanks for the feedback, when I make an update I will re-post it here.


(Ruber Eaglenest) #4

Yeah, play in the family, with friends, the mobile speaking and everybody shouting the actions. Yeah, it is not ideal, but that’s how we played last time :wink:

About options for the Voice2Text, I can think of only two right now 1) automatic, as it is now, and 2) a press to speak option. That is, that the voice recognition is always paused until the player hits the screen.

Regards!


#5

Cool!! I will create a settings page where you can choose if you want the Voice2Text to be automatic or not, and also a button (or textfield) to add text in audio mode like you suggested. I hope to release an update until the end of the month!
Thanks again for the feedback!

Regards!


#6

Hi guys, i made an update where you can type text on audio mode for the difficult words by long pressing the screen! Its on google play.

https://play.google.com/store/apps/details?id=org.redrossistudios.audioif

Regards,
Daniel


(Parser Commander) #7

Nice work, Daniel!

And great idea, Ruber, to play in group! :smiley:

I see good business there selling family games… ^.^

I was testing the software with a game in Spanish without looking at the screen, and it works very well. I have just one suggestion.

Sometimes the software misunderstood a word, or I speak too quickly before the system was ready to hear, so i send a wrong command to the game. Then the game says that this can’t be done, driving me to think that what i said can’t really be done, when the problem was that it was an “error”. When this happens in a written game it’s easy to see misspell in the input, but with only voice without looking at the screen there is no way to know that.

Perhaps there could be an option that makes the software repeat the input I say, so i can ensure it understood what I said. It could be a bit tedious to hear the same every time, but i can’t imagine another way to deal with this issue. Perhaps the repeated command could be said by another system voice (masculine if the main one is feminine), and do it at lower volume, while the main voice continues narrating the normal response. Just ideas, but not sure what could work better.

Anyway, nice experience to play that way. This can be a good sport training copanion :smiley: