Use FAST FAST FAST arpabet 2.0 crate

Cargo.lock

@@ -335,7 +335,40 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 
 [[package]]
 name = "arpabet"
-version = "1.0.0"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "arpabet_cmudict 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "arpabet_parser 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "arpabet_types 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
+[[package]]
+name = "arpabet_cmudict"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "arpabet_parser 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "arpabet_types 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "lazy_static 1.4.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "phf 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "phf_codegen 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "regex 1.3.7 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
+[[package]]
+name = "arpabet_parser"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "arpabet_types 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "lazy_static 1.4.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "regex 1.3.7 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
+[[package]]
+name = "arpabet_types"
+version = "2.0.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 dependencies = [
  "lazy_static 1.4.0 (registry+https://github.com/rust-lang/crates.io-index)",
@@ -901,10 +934,10 @@ dependencies = [
 
 [[package]]
 name = "grapheme_to_phoneme"
-version = "0.0.3"
+version = "0.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 dependencies = [
- "arpabet 1.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "arpabet 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "ndarray 0.13.1 (registry+https://github.com/rust-lang/crates.io-index)",
  "ndarray-npy 0.5.0 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
@@ -1467,6 +1500,15 @@ dependencies = [
  "proc-macro-hack 0.5.11 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
+[[package]]
+name = "phf_codegen"
+version = "0.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "phf_generator 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "phf_shared 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
 [[package]]
 name = "phf_generator"
 version = "0.8.0"
@@ -2222,13 +2264,13 @@ dependencies = [
  "actix-utils 1.0.6 (registry+https://github.com/rust-lang/crates.io-index)",
  "actix-web 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "anyhow 1.0.30 (registry+https://github.com/rust-lang/crates.io-index)",
- "arpabet 1.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "arpabet 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "base64 0.12.2 (registry+https://github.com/rust-lang/crates.io-index)",
  "chrono 0.4.11 (registry+https://github.com/rust-lang/crates.io-index)",
  "diesel 1.4.4 (registry+https://github.com/rust-lang/crates.io-index)",
  "env_logger 0.7.1 (registry+https://github.com/rust-lang/crates.io-index)",
  "futures 0.1.29 (registry+https://github.com/rust-lang/crates.io-index)",
- "grapheme_to_phoneme 0.0.3 (registry+https://github.com/rust-lang/crates.io-index)",
+ "grapheme_to_phoneme 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "hostname 0.3.1 (registry+https://github.com/rust-lang/crates.io-index)",
  "hound 3.4.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "limitation 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)",
@@ -2472,7 +2514,10 @@ dependencies = [
 "checksum aho-corasick 0.7.6 (registry+https://github.com/rust-lang/crates.io-index)" = "58fb5e95d83b38284460a5fda7d6470aa0b8844d283a0b614b8535e880800d2d"
 "checksum anyhow 1.0.30 (registry+https://github.com/rust-lang/crates.io-index)" = "2494382e9ba43995f3c56359e518641f450f5c36feeb4632a75cde2ec297c867"
 "checksum arc-swap 0.4.4 (registry+https://github.com/rust-lang/crates.io-index)" = "d7b8a9123b8027467bce0099fe556c628a53c8d83df0507084c31e9ba2e39aff"
-"checksum arpabet 1.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "c76d18638cfaa15c3e7c521ef8a223614bee6b802433d788815347f28a0c4add"
+"checksum arpabet 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "4d7bcdb435d536dfa262abef687e3e0517da921cb56bda1d56c6a03e1d7b8d16"
+"checksum arpabet_cmudict 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "493246593f4f5b249ea42ff224e2e60655a305762959e45c788cca684b7b446f"
+"checksum arpabet_parser 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "e1a5b409d512e07dfabce3cdd864470fa6bf51e8aa762eeee3265e359f0687a9"
+"checksum arpabet_types 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "ae3a6e204a66085d5d06c76b16bb2218d496ee291a5a0fad221a9c53b08862e4"
 "checksum ascii 0.9.3 (registry+https://github.com/rust-lang/crates.io-index)" = "eab1c04a571841102f5345a8fc0f6bb3d31c315dec879b5c6e42e40ce7ffa34e"
 "checksum async-trait 0.1.36 (registry+https://github.com/rust-lang/crates.io-index)" = "a265e3abeffdce30b2e26b7a11b222fe37c6067404001b434101457d0385eb92"
 "checksum atty 0.2.14 (registry+https://github.com/rust-lang/crates.io-index)" = "d9b39be18770d11421cdb1b9947a45dd3f37e93092cbf377614828a319d5fee8"
@@ -2538,7 +2583,7 @@ dependencies = [
 "checksum fxhash 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)" = "c31b6d751ae2c7f11320402d34e41349dd1016f8d5d45e48c4312bc8625af50c"
 "checksum generic-array 0.12.3 (registry+https://github.com/rust-lang/crates.io-index)" = "c68f0274ae0e023facc3c97b2e00f076be70e254bc851d972503b328db79b2ec"
 "checksum getrandom 0.1.13 (registry+https://github.com/rust-lang/crates.io-index)" = "e7db7ca94ed4cd01190ceee0d8a8052f08a247aa1b469a7f68c6a3b71afcf407"
-"checksum grapheme_to_phoneme 0.0.3 (registry+https://github.com/rust-lang/crates.io-index)" = "df2915a3acab6ce0b29f2f88f63f5c53d67c7335705355e4ff017245bd4d83d3"
+"checksum grapheme_to_phoneme 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "a1af33031ec99fd5585783e2caba8e017c01a9efb1318fed9fc9250420dd019f"
 "checksum h2 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)" = "b9433d71e471c1736fd5a61b671fc0b148d7a2992f666c958d03cd8feb3b88d1"
 "checksum heck 0.3.1 (registry+https://github.com/rust-lang/crates.io-index)" = "20564e78d53d2bb135c343b3f47714a56af2061f1c928fdb541dc7b9fdd94205"
 "checksum hermit-abi 0.1.6 (registry+https://github.com/rust-lang/crates.io-index)" = "eff2656d88f158ce120947499e971d743c05dbcbed62e5bd2f38f1698bbc3772"
@@ -2603,6 +2648,7 @@ dependencies = [
 "checksum pest_generator 2.1.3 (registry+https://github.com/rust-lang/crates.io-index)" = "99b8db626e31e5b81787b9783425769681b347011cc59471e33ea46d2ea0cf55"
 "checksum pest_meta 2.1.3 (registry+https://github.com/rust-lang/crates.io-index)" = "54be6e404f5317079812fc8f9f5279de376d8856929e21c184ecf6bbd692a11d"
 "checksum phf 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)" = "3dfb61232e34fcb633f43d12c58f83c1df82962dcdfa565a4e866ffc17dafe12"
+"checksum phf_codegen 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)" = "cbffee61585b0411840d3ece935cce9cb6321f01c45477d30066498cd5e1a815"
 "checksum phf_generator 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)" = "17367f0cc86f2d25802b2c26ee58a7b23faeccf78a396094c13dced0d0182526"
 "checksum phf_macros 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)" = "7f6fde18ff429ffc8fe78e2bf7f8b7a5a5a6e2a8b58bc5a9ac69198bbda9189c"
 "checksum phf_shared 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)" = "c00cf8b9eafe68dde5e9eaa2cef8ee84a9336a47d566ec55ca16589633b65af7"

tts_service/Cargo.toml

@@ -40,7 +40,7 @@ toml = "0.5.6"
 #rusty_redis_rate_limiting = { git = "https://github.com/tinrab/rusty-redis-rate-limiting", rev = "46d95040708df92d46cef22d319d0182e12a2c0c" }
 
 # My libraries
-arpabet = "1.0.0"
-grapheme_to_phoneme = "0.0.3"
+arpabet = "2.0.0"
+grapheme_to_phoneme = "0.1.0"
 numerics = "0.0.1"
 sentence = "0.0.2"

tts_service/deployed_models.toml

@@ -17,7 +17,7 @@ name = "Moistcr1tikal"
 slug = "moistcr1tikal"
 model_pipeline = "ArpabetGlowTtsMelgan"
 glow_tts = "glow-tts_moistcr1tikal_ljstx_2020.10.09_expr-1_chkpt-4858.torchjit"
-melgan = "melgan_manyvoice5.0_2020-07-23_12d5838_10760.torchjit" # TODO: Needs custom vocoder
+melgan = "melgan_moistcr1tikal_2020-10-10_d3b9647_12000.torchjit" # TODO: Train more
 
 [[speakers]]
 name = "Paula Deen"

tts_service/models.toml

@@ -43,7 +43,7 @@ name = "Moistcr1tikal"
 slug = "moistcr1tikal"
 model_pipeline = "ArpabetGlowTtsMelgan"
 glow_tts = "glow-tts_moistcr1tikal_ljstx_2020.10.09_expr-1_chkpt-4858.torchjit"
-melgan = "melgan_manyvoice5.0_2020-07-23_12d5838_10760.torchjit" # TODO: Needs custom vocoder
+melgan = "melgan_moistcr1tikal_2020-10-10_d3b9647_12000.torchjit" # TODO: Train more
 
 [[speakers]]
 name = "Paula Deen"

tts_service/src/endpoints/speak/legacy_tts.rs

@@ -8,7 +8,7 @@ use actix_web::{
   HttpResponse,
 };
 
-use arpabet::Arpabet;
+use arpabet;
 use crate::AppState;
 use crate::model::old_model::TacoMelModel;
 use crate::text::arpabet::text_to_arpabet_encoding;
@@ -48,8 +48,8 @@ pub async fn post_tts(_request: HttpRequest,
 
   let cleaned_text = clean_text(&text);
 
-  let arpabet = Arpabet::load_cmudict();
-  let encoded = text_to_arpabet_encoding(arpabet, &cleaned_text);
+  let cmudict = arpabet::load_cmudict();
+  let encoded = text_to_arpabet_encoding(cmudict, &cleaned_text);
 
   let sample_rate_hz = app_state.default_sample_rate_hz;
 

tts_service/src/endpoints/speak/speak.rs

@@ -9,7 +9,7 @@ use actix_web::{
   Result as ActixResult,
 };
 
-use arpabet::Arpabet;
+use arpabet;
 use crate::AppState;
 use crate::database::model::NewSentence;
 use crate::endpoints::helpers::ip_address::get_request_ip;
@@ -78,8 +78,8 @@ pub async fn post_speak(request: HttpRequest,
       info!("Melgan Model: {}", melgan_model);
       info!("Text: {}", text);
 
-      let arpabet = Arpabet::load_cmudict();
-      let encoded = text_to_arpabet_encoding(arpabet, &cleaned_text);
+      let cmudict = arpabet::load_cmudict();
+      let encoded = text_to_arpabet_encoding(cmudict, &cleaned_text);
 
       let tacotron = app_state.model_cache.get_or_load_arbabet_tacotron(&tacotron_model)
           .expect(&format!("Couldn't load tacotron model: {}", tacotron_model));

tts_service/src/main.rs

@@ -32,6 +32,8 @@ use anyhow::Result as AnyhowResult;
 
 use actix::{Addr, SyncArbiter, Context};
 use arpabet::Arpabet;
+use arpabet::load_cmudict;
+use arpabet::load_from_file as arpabet_load_from_file;
 use crate::database::connector::DatabaseConnector;
 use crate::database::sentence_recorder::SentenceRecorder;
 use crate::endpoints::index::get_root;
@@ -295,15 +297,15 @@ pub fn main() -> AnyhowResult<()> {
       info!("Loading default CMUdict Arpabet...");
       // Here we introduce silly internet words such as "lulz".
       // NB: The extras file takes precedence for any dictionary collisions.
-      let cmu_dict = Arpabet::load_cmudict();
+      let cmu_dict = load_cmudict();
       info!("Loading Arpabet extensions...");
-      let extra_arpabet = Arpabet::load_from_file(&extras_filename)?;
+      let extra_arpabet = arpabet_load_from_file(&extras_filename)?;
       let arpabet = cmu_dict.combine(&extra_arpabet);
       arpabet.clone()
     },
     None => {
       info!("Loading default CMUdict Arpabet...");
-      Arpabet::load_cmudict().clone()
+      load_cmudict().clone()
     },
   };