I think this would be great. Take off the shelf (university GNU parsing software using semantics and syntax), combine it with off the shelf speech recognition software (look into Sphinx) and command your troops using this technology. Has anyone ever read the transcripts produced by SHRDLU? SHRDLU dates to 1970 and used 64k. It is very achievable intelligence when confined to a domain such as SHRDLU''s. RTSs have domains which are perfectly suitable to such engines. Command your troops via speech recognition: YOU: "Royce, move your troops over the hill and wait." COMMANDER: "Roger that..." YOU: "Jensen, What is the situation with you guys?" JENSEN: "Sir, we''re holed up under enemy fire. We need reinforcements." YOU: "Baker, status?" BAKER: "The north drainage is providing cover for us at the moment. We''re moving north, ETA to the pass is 1 hour." YOU: "Why are you in the north drainage?" BAKER: "Sir, the south drainage was too exposed." YOU: "Johnson, I need air coverage for the 5th battalion at cooridnates... Enemy approaching from the west." ------ Details: Let''s look at the first example. You say: "Royce, move your troops over the hill and wait." The parser identifies Royce as interprets you message to be directed to Royce. ''Move'' is a verb and the parser interprets it as a command. ''Your troops'' is a noun phrase, ''over'' is a preposition, ''the hill'' is a noun phrase (a direct object) and ''wait is a successive command. ''The hill'' is ambiguous, but context should identify which hill is being discussed. Look into SHRDLU. You will be amazed at what is possible when limited domains are used. Remember, SHRDLU was programmed in 1970 and had a conceptual understanding of what was being discussed, and had knowledge and memory of discourse. Use Google.

A SHRDLU link for the lazy.

Another SHRDLU link for even more lazy.

Wow... I just skimmed it, but... So much potential!

I know you intend this to be built directly into the game, but using Dragon with Starcraft can achieve much the same results.

ld

quote: Original post by liquiddark
I know you intend this to be built directly into the game, but using Dragon with Starcraft can achieve much the same results.

I haven''t used either, but I don''t believe you. I don''t for one minute believe there is any game out there that has anything remotely close to the true discourse modeling exhinited by SHRDLU.

Not even close.

I could be wrong, because I haven''t played Starcraft.

But I would venture that I am not wrong.

Care to port it from MacLisp?

[ GDNet Start Here | GDNet Search Tool | GDNet FAQ | MS RTFM [MSDN] | SGI STL Docs | Google! ]
Thanks to Kylotan for the idea!

quote: Original post by bishop_pass
I don''t for one minute believe there is any game out there that has anything remotely close to the true discourse modeling exhinited by SHRDLU.

You''re right. However, it doesn''t matter. Dragon does the job nicely, and you can teach it enough so that giving orders is well-handled.

Which is my point.

ld

SHRDLU....never heard of it before now, but that transcript between the user and the computer on the first link you provided was amazing, especially considering the memory constraints.

I would love to see a game like the one bishop_pass suggested as I have more than a passing interest in strategy and tactics games.....oooh my mouth is watering at the prospect.

- DarkIce

Holy shit. I tend to be a little bit skeptical of non-standard interfaces, but that is pretty awesome, and oozes unused potential. Anyone fancy porting it to C++?

Everyone, be sure to read the transcript slowly and all the way through. The second link provides commentary along the way. Especially note SHRDLU''s internal modeling of the world and past events. SHRDLU explains what it did in the past, why it did it, and what the situation was in the past, in addition to the present. SHRDLU disagrees with the user when the user implies something untrue. New concepts are exlplained to SHRDLU, such as ''steeple'' and then SHRDLU proceeds to work with steeples as though SHRDLU always knew what they were.

Admittiedly, the world is simple, but look at the complexity of the sentences which SHRDLU understands, and how SHRDLU disambiguates very complex sentences.