add cmudict dataset

.gitignore

@@ -4,4 +4,4 @@
 docs/build/
 docs/site/
 docs/flux.css
-demos
+deps

src/Flux.jl

@@ -29,4 +29,6 @@ include("layers/basic.jl")
 include("layers/recurrent.jl")
 include("layers/normalisation.jl")
 
+include("data/Data.jl")
+
 end # module

src/data/Data.jl

@@ -0,0 +1,14 @@
+module Data
+
+export CMUDict, cmudict
+
+deps(path...) = joinpath(@__DIR__, "..", "..", "deps", path...)
+
+function __init__()
+  mkpath(deps())
+end
+
+include("cmudict.jl")
+using .CMUDict
+
+end

src/data/cmudict.jl

@@ -0,0 +1,42 @@
+module CMUDict
+
+export cmudict
+
+using ..Data: deps
+
+const version = "0.7b"
+
+function load()
+  isdir(deps("cmudict")) && return
+  mkpath(deps("cmudict"))
+  for x in ["", ".phones", ".symbols"]
+    download("http://svn.code.sf.net/p/cmusphinx/code/trunk/cmudict/cmudict-$version$x",
+             deps("cmudict", "cmudict$x"))
+  end
+end
+
+function phones()
+  load()
+  Symbol.(first.(split.(split(readstring(deps("cmudict", "cmudict.phones")),
+                        "\n", keep = false), "\t")))
+end
+
+function symbols()
+  load()
+  Symbol.(split(readstring(deps("CMUDict", "cmudict.symbols")),
+                "\n", keep = false))
+end
+
+function rawdict()
+  load()
+  Dict(String(xs[1]) => Symbol.(xs[2:end]) for xs in
+       filter(!isempty, split.(split(readstring(deps("CMUDict", "cmudict")), "\n"))))
+end
+
+validword(s) = ismatch(r"^[\w-\.]+$", s)
+
+cmudict() = filter((s, ps) -> validword(s), rawdict())
+
+alphabet() = ['A':'Z'..., '0':'9'..., '_', '-', '.']
+
+end

test/data.jl

@@ -0,0 +1,3 @@
+using Flux.Data
+
+@test cmudict()["CATASTROPHE"] == :[K,AH0,T,AE1,S,T,R,AH0,F,IY0].args