Chinese CVV: custom dictionary support

Author: oxygen-dioxide

OpenUtau.Plugin.Builtin/ChineseCVVMonophonePhonemizer.cs

@@ -1,16 +1,62 @@
+using System;
 using System.Collections.Generic;
+using System.IO;
 using System.Linq;
+using System.Threading.Tasks;
 using OpenUtau.Api;
-using OpenUtau.Core;
-using OpenUtau.Core.Ustx;
+using Serilog;
 
-namespace OpenUtau.Plugin.Builtin {
+namespace OpenUtau.Plugin.Builtin
+{
     /// <summary>
     /// Chinese 十月式整音扩张 CVV Phonemizer.
     /// <para>It works by spliting "duang" to "duang" + "_ang", to produce the proper tail sound.</para>
     /// </summary>
     [Phonemizer("Chinese CVV (十月式整音扩张) Phonemizer", "ZH CVV", language: "ZH")]
-    public class ChineseCVVPhonemizer : BaseChinesePhonemizer {
+    public class ChineseCVVMonophonePhonemizer : MonophonePhonemizer
+    {
+        static readonly string pinyins = "a,ai,an,ang,ao,ba,bai,ban,bang,bao,bei,ben,beng,bi,bian,biao,bie,bin,bing,bo,bu,ca,cai,can,cang,cao,ce,cei,cen,ceng,cha,chai,chan,chang,chao,che,chen,cheng,chi,chong,chou,chu,chua,chuai,chuan,chuang,chui,chun,chuo,ci,cong,cou,cu,cuan,cui,cun,cuo,da,dai,dan,dang,dao,de,dei,den,deng,di,dia,dian,diao,die,ding,diu,dong,dou,du,duan,dui,dun,duo,e,ei,en,eng,er,fa,fan,fang,fei,fen,feng,fo,fou,fu,ga,gai,gan,gang,gao,ge,gei,gen,geng,gong,gou,gu,gua,guai,guan,guang,gui,gun,guo,ha,hai,han,hang,hao,he,hei,hen,heng,hong,hou,hu,hua,huai,huan,huang,hui,hun,huo,ji,jia,jian,jiang,jiao,jie,jin,jing,jiong,jiu,ju,jv,juan,jvan,jue,jve,jun,jvn,ka,kai,kan,kang,kao,ke,kei,ken,keng,kong,kou,ku,kua,kuai,kuan,kuang,kui,kun,kuo,la,lai,lan,lang,lao,le,lei,leng,li,lia,lian,liang,liao,lie,lin,ling,liu,lo,long,lou,lu,luan,lun,luo,lv,lve,ma,mai,man,mang,mao,me,mei,men,meng,mi,mian,miao,mie,min,ming,miu,mo,mou,mu,na,nai,nan,nang,nao,ne,nei,nen,neng,ni,nian,niang,niao,nie,nin,ning,niu,nong,nou,nu,nuan,nun,nuo,nv,nve,o,ou,pa,pai,pan,pang,pao,pei,pen,peng,pi,pian,piao,pie,pin,ping,po,pou,pu,qi,qia,qian,qiang,qiao,qie,qin,qing,qiong,qiu,qu,qv,quan,qvan,que,qve,qun,qvn,ran,rang,rao,re,ren,reng,ri,rong,rou,ru,rua,ruan,rui,run,ruo,sa,sai,san,sang,sao,se,sen,seng,sha,shai,shan,shang,shao,she,shei,shen,sheng,shi,shou,shu,shua,shuai,shuan,shuang,shui,shun,shuo,si,song,sou,su,suan,sui,sun,suo,ta,tai,tan,tang,tao,te,tei,teng,ti,tian,tiao,tie,ting,tong,tou,tu,tuan,tui,tun,tuo,wa,wai,wan,wang,wei,wen,weng,wo,wu,xi,xia,xian,xiang,xiao,xie,xin,xing,xiong,xiu,xu,xv,xuan,xvan,xue,xve,xun,xvn,ya,yan,yang,yao,ye,yi,yin,ying,yo,yong,you,yu,yv,yuan,yvan,yue,yve,yun,yvn,za,zai,zan,zang,zao,ze,zei,zen,zeng,zha,zhai,zhan,zhang,zhao,zhe,zhei,zhen,zheng,zhi,zhong,zhou,zhu,zhua,zhuai,zhuan,zhuang,zhui,zhun,zhuo,zi,zong,zou,zu,zuan,zui,zun";
+        static readonly string tails = "_vn,_ing,_ong,_an,_ou,_er,_ao,_eng,_ang,_en,_en2,_ai,_iong,_in,_ei";
+        
+        static readonly string[] pinyinList = pinyins.Split(',');
+        static readonly string[] tailList = tails.Split(',');
+
+        public ChineseCVVMonophonePhonemizer() {
+            ConsonantLength = 120;    
+        }
+
+        protected override IG2p LoadG2p() {
+            var g2ps = new List<IG2p>();
+
+            // Load dictionary from plugin folder.
+            string path = Path.Combine(PluginDir, "zhcvv.yaml");
+            if (File.Exists(path)) {
+                g2ps.Add(G2pDictionary.NewBuilder().Load(File.ReadAllText(path)).Build());
+            }
+
+            // Load dictionary from singer folder.
+            if (singer != null && singer.Found && singer.Loaded) {
+                string file = Path.Combine(singer.Location, "zhcvv.yaml");
+                if (File.Exists(file)) {
+                    try {
+                        g2ps.Add(G2pDictionary.NewBuilder().Load(File.ReadAllText(file)).Build());
+                    } catch (Exception e) {
+                        Log.Error(e, $"Failed to load {file}");
+                    }
+                }
+            }
+            g2ps.Add(new ChineseCVVG2p());
+            return new G2pFallbacks(g2ps.ToArray());
+        }
+
+        protected override Dictionary<string, string[]> LoadVowelFallbacks() {
+            return "_un=_en".Split(';')
+                .Select(entry => entry.Split('='))
+                .ToDictionary(parts => parts[0], parts => parts[1].Split(','));
+        }
+    }
+    
+    class ChineseCVVG2p : IG2p{
         /// <summary>
         ///  The consonant table.
         /// </summary>
@@ -22,26 +68,22 @@ public class ChineseCVVPhonemizer : BaseChinesePhonemizer {
 
         static HashSet<string> cSet;
         static Dictionary<string, string> vDict;
-
-        static ChineseCVVPhonemizer() {
+        
+        static ChineseCVVG2p() {
             cSet = new HashSet<string>(consonants.Split(','));
             vDict = vowels.Split(',')
                 .Select(s => s.Split('='))
                 .ToDictionary(a => a[0], a => a[1]);
         }
 
-        private USinger singer;
-
-        // Simply stores the singer in a field.
-        public override void SetSinger(USinger singer) => this.singer = singer;
+        public bool IsVowel(string phoneme){
+            return !phoneme.StartsWith("_");
+        }
 
-        public override Result Process(Note[] notes, Note? prev, Note? next, Note? prevNeighbour, Note? nextNeighbour, Note[] prevNeighbours) {
+        public string[] Query(string lyric){
             // The overall logic is:
             // 1. Remove consonant: "duang" -> "uang".
             // 2. Lookup the trailing sound in vowel table: "uang" -> "_ang".
-            // 3. Split the total duration and returns "duang" and "_ang".
-            var lyric = notes[0].lyric;
-            var note = notes[0];
             string consonant = string.Empty;
             string vowel = string.Empty;
             if (lyric.Length > 2 && cSet.Contains(lyric.Substring(0, 2))) {
@@ -63,62 +105,20 @@ public override Result Process(Note[] notes, Note? prev, Note? next, Note? prevN
             if ((vowel == "an") && (consonant == "y")) {
                 vowel = "ian";
             }
-            string phoneme0 = lyric;
-            // Get color
-            string color = string.Empty;
-            int toneShift = 0;
-            if (note.phonemeAttributes != null) {
-                var attr = note.phonemeAttributes.FirstOrDefault(attr => attr.index == 0);
-                color = attr.voiceColor;
-                toneShift = attr.toneShift;
+            if(vDict.TryGetValue(vowel, out var tail)){
+                return new string[] { lyric, tail };
+            }else{
+                return new string[] { lyric };
             }
-            // We will need to split the total duration for phonemes, so we compute it here.
-            int totalDuration = notes.Sum(n => n.duration);
-            // Lookup the vowel split table. For example, "uang" will match "_ang".
-            if (vDict.TryGetValue(vowel, out var phoneme1)) {
-                // Now phoneme0="duang" and phoneme1="_ang",
-                // try to give "_ang" 120 ticks, but no more than half of the total duration.
-                int length1 = 120;
-                if (length1 > totalDuration / 2) {
-                    length1 = totalDuration / 2;
-                }
-                if (singer.TryGetMappedOto(phoneme0, note.tone + toneShift, color, out var oto0)) {
-                    phoneme0 = oto0.Alias;
-                }
-
-                if (singer.TryGetMappedOto(phoneme1, note.tone + toneShift, color, out var oto1)) {
-                    phoneme1 = oto1.Alias;
-                }
-
-                if (phoneme1.Contains("_un") && !singer.TryGetMappedOto(phoneme1, note.tone + toneShift, color, out var oto2)) {
-                    phoneme1 = "_en";
-                } else if (phoneme1.Contains("_un") && singer.TryGetMappedOto(phoneme1, note.tone + toneShift, color, out var oto3)) {
-                    phoneme1 = oto3.Alias;
-                }
+                        
+        }
+        public bool IsValidSymbol(string symbol){
+            return true;
+        }
 
-                return new Result {
-                    phonemes = new Phoneme[] {
-                        new Phoneme() {
-                            phoneme = phoneme0,
-                        },
-                        new Phoneme() {
-                            phoneme = phoneme1,
-                            position = totalDuration - length1,
-                        }
-                    },
-                };
-            }
-            if (singer.TryGetMappedOto(phoneme0, note.tone + toneShift, color, out var oto)) {
-                phoneme0 = oto.Alias;
-            }
-            // Not spliting is needed. Return as is.
-            return new Result {
-                phonemes = new Phoneme[] {
-                    new Phoneme() {
-                        phoneme = phoneme0,
-                    }
-                },
-            };
+        public string[] UnpackHint(string hint, char separator = ' ') {
+            return hint.Split(separator)
+                .ToArray();
         }
     }
-}
+}