Amkor/ENIG
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

add supertonic

Browse Source
This commit is contained in:
backuppc
2025-12-10 11:39:15 +09:00
parent 3695ab0044
commit 868fa2deec
22 changed files with 54223 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

69
Cs_HMI/AGVLogic/AGVNavigationCore/Controls/UnifiedAGVCanvas.Events.cs
View File

@@ -83,6 +83,75 @@ namespace AGVNavigationCore.Controls
// UI 정보 그리기 (변환 없이) // UI 정보 그리기 (변환 없이)
if (_showGrid) if (_showGrid)
DrawUIInfo(g); DrawUIInfo(g);
// 동기화 화면 그리기 (변환 없이, 최상위)
if (_canvasMode == CanvasMode.Sync)
{
DrawSyncScreen(g);
}
}
private void DrawSyncScreen(Graphics g)
{
// 반투명 검은색 배경
using (var brush = new SolidBrush(Color.FromArgb(200, 0, 0, 0)))
{
g.FillRectangle(brush, this.ClientRectangle);
}
// 중앙에 메시지 표시
var center = new Point(Width / 2, Height / 2);
// 메시지 폰트
using (var fontTitle = new Font("Malgun Gothic", 24, FontStyle.Bold))
using (var fontDetail = new Font("Malgun Gothic", 14))
using (var brushText = new SolidBrush(Color.White))
{
// 메인 메시지
var sizeTitle = g.MeasureString(_syncMessage, fontTitle);
g.DrawString(_syncMessage, fontTitle, brushText,
center.X - sizeTitle.Width / 2,
center.Y - sizeTitle.Height / 2 - 60);
// 진행률 바 배경
int barWidth = 500;
int barHeight = 30;
int barX = center.X - barWidth / 2;
int barY = center.Y + 10;
using (var brushBarBg = new SolidBrush(Color.FromArgb(64, 64, 64)))
{
g.FillRectangle(brushBarBg, barX, barY, barWidth, barHeight);
}
g.DrawRectangle(Pens.Gray, barX, barY, barWidth, barHeight);
// 진행률 바 채우기
if (_syncProgress > 0)
{
using (var brushProgress = new SolidBrush(Color.LimeGreen))
{
int fillWidth = (int)((barWidth - 4) * _syncProgress);
if (fillWidth > 0)
g.FillRectangle(brushProgress, barX + 2, barY + 2, fillWidth, barHeight - 4);
}
}
// 진행률 텍스트
string progressText = $"{(_syncProgress * 100):F0}%";
var sizeProgress = g.MeasureString(progressText, fontDetail);
g.DrawString(progressText, fontDetail, brushText,
center.X - sizeProgress.Width / 2,
barY + 5);
// 상세 메시지
if (!string.IsNullOrEmpty(_syncDetail))
{
var sizeDetail = g.MeasureString(_syncDetail, fontDetail);
g.DrawString(_syncDetail, fontDetail, brushText,
center.X - sizeDetail.Width / 2,
barY + barHeight + 20);
}
}
} }
private void DrawGrid(Graphics g) private void DrawGrid(Graphics g)

42
Cs_HMI/AGVLogic/AGVNavigationCore/Controls/UnifiedAGVCanvas.cs
View File

@@ -35,7 +35,8 @@ namespace AGVNavigationCore.Controls
/// </summary> /// </summary>
public enum CanvasMode public enum CanvasMode
{ {
Edit // 편집 가능 (맵 에디터) Edit, // 편집 가능 (맵 에디터)
Sync // 동기화 모드 (장비 설정 동기화)
} }
/// <summary> /// <summary>
@@ -116,6 +117,11 @@ namespace AGVNavigationCore.Controls
// RFID 중복 검사 // RFID 중복 검사
private HashSet<string> _duplicateRfidNodes = new HashSet<string>(); private HashSet<string> _duplicateRfidNodes = new HashSet<string>();
// 동기화 모드 관련
private string _syncMessage = "동기화 중...";
private float _syncProgress = 0.0f;
private string _syncDetail = "";
// 브러쉬 및 펜 // 브러쉬 및 펜
private Brush _normalNodeBrush; private Brush _normalNodeBrush;
private Brush _rotationNodeBrush; private Brush _rotationNodeBrush;
@@ -546,6 +552,40 @@ namespace AGVNavigationCore.Controls
} }
} }
/// <summary>
/// 동기화 상태 설정
/// </summary>
/// <param name="message">메인 메시지</param>
/// <param name="progress">진행률 (0.0 ~ 1.0)</param>
/// <param name="detail">상세 메시지</param>
public void SetSyncStatus(string message, float progress, string detail = "")
{
_syncMessage = message;
_syncProgress = Math.Max(0.0f, Math.Min(1.0f, progress));
_syncDetail = detail;
if (_canvasMode != CanvasMode.Sync)
{
_canvasMode = CanvasMode.Sync;
UpdateModeUI();
}
Invalidate();
}
/// <summary>
/// 동기화 모드 종료
/// </summary>
public void ExitSyncMode()
{
if (_canvasMode == CanvasMode.Sync)
{
_canvasMode = CanvasMode.Edit; // 기본 모드로 복귀 (또는 이전 모드)
UpdateModeUI();
Invalidate();
}
}
#endregion #endregion
#region Cleanup #region Cleanup

12
Cs_HMI/Project/StateMachine/Step/_SM_RUN_SYNC.cs
View File

@@ -107,6 +107,13 @@ namespace Project
var item = synlist.ElementAt(synidx); var item = synlist.ElementAt(synidx);
UpdateProgressStatus(stepTime.TotalSeconds, 5, $"SYNC :{item.Key}"); UpdateProgressStatus(stepTime.TotalSeconds, 5, $"SYNC :{item.Key}");
PUB.AGV.AGVCommand(item.Key, item.Value); PUB.AGV.AGVCommand(item.Key, item.Value);
// 캔버스에 동기화 상태 표시
if (PUB._mapCanvas != null)
{
float progress = (float)synidx / synlist.Count;
PUB._mapCanvas.SetSyncStatus("장비 설정 동기화 중...", progress, $"항목: {item.Key} ({synidx + 1}/{synlist.Count})");
}
} }
LastCommandTime = DateTime.Now; LastCommandTime = DateTime.Now;
PUB.sm.UpdateRunStepSeq(); PUB.sm.UpdateRunStepSeq();
@@ -143,6 +150,11 @@ namespace Project
{ {
PUB.AddEEDB($"SYNC완료({PUB.Result.TargetPos})"); PUB.AddEEDB($"SYNC완료({PUB.Result.TargetPos})");
UpdateProgressStatus(stepTime.TotalSeconds, 5, "SYNC : 완료"); UpdateProgressStatus(stepTime.TotalSeconds, 5, "SYNC : 완료");
// 동기화 완료 시 캔버스 모드 복귀
if (PUB._mapCanvas != null)
PUB._mapCanvas.SetSyncStatus("동기화 완료!", 1.0f, "잠시 후 메인 화면으로 이동합니다.");
LastCommandTime = DateTime.Now; LastCommandTime = DateTime.Now;
PUB.sm.UpdateRunStepSeq(); PUB.sm.UpdateRunStepSeq();
return false; return false;

4
Cs_HMI/Project/StateMachine/_Loop.cs
View File

@@ -156,6 +156,10 @@ namespace Project
// 장치 관리 태스크 시작 (IDLE 진입 시 한 번만) // 장치 관리 태스크 시작 (IDLE 진입 시 한 번만)
StartDeviceManagementTask(); StartDeviceManagementTask();
// 동기화 모드 종료 (혹시 남아있을 경우)
if (PUB._mapCanvas != null)
PUB._mapCanvas.ExitSyncMode();
} }
//자동소거기능 //자동소거기능

135
Cs_HMI/SubProject/SuperTonic/ExampleONNX.cs Normal file
View File

@@ -0,0 +1,135 @@
using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
namespace Supertonic
{
class Program
{
class Args
{
public bool UseGpu { get; set; } = false;
public string OnnxDir { get; set; } = "assets/onnx";
public int TotalStep { get; set; } = 5;
public float Speed { get; set; } = 1.05f;
public int NTest { get; set; } = 4;
public List<string> VoiceStyle { get; set; } = new List<string> { "assets/voice_styles/M1.json" };
public List<string> Text { get; set; } = new List<string>
{
"This morning, I took a walk in the park, and the sound of the birds and the breeze was so pleasant that I stopped for a long time just to listen."
};
public string SaveDir { get; set; } = "results";
public bool Batch { get; set; } = false;
}
static Args ParseArgs(string[] args)
{
var result = new Args();
for (int i = 0; i < args.Length; i++)
{
switch (args[i])
{
case "--use-gpu":
result.UseGpu = true;
break;
case "--batch":
result.Batch = true;
break;
case "--onnx-dir" when i + 1 < args.Length:
result.OnnxDir = args[++i];
break;
case "--total-step" when i + 1 < args.Length:
result.TotalStep = int.Parse(args[++i]);
break;
case "--speed" when i + 1 < args.Length:
result.Speed = float.Parse(args[++i]);
break;
case "--n-test" when i + 1 < args.Length:
result.NTest = int.Parse(args[++i]);
break;
case "--voice-style" when i + 1 < args.Length:
result.VoiceStyle = args[++i].Split(',').ToList();
break;
case "--text" when i + 1 < args.Length:
result.Text = args[++i].Split('|').ToList();
break;
case "--save-dir" when i + 1 < args.Length:
result.SaveDir = args[++i];
break;
}
}
return result;
}
static void Main(string[] args)
{
Console.WriteLine("=== TTS Inference with ONNX Runtime (C#) ===\n");
// --- 1. Parse arguments --- //
var parsedArgs = ParseArgs(args);
int totalStep = parsedArgs.TotalStep;
float speed = parsedArgs.Speed;
int nTest = parsedArgs.NTest;
string saveDir = parsedArgs.SaveDir;
var voiceStylePaths = parsedArgs.VoiceStyle;
var textList = parsedArgs.Text;
bool batch = parsedArgs.Batch;
if (voiceStylePaths.Count != textList.Count)
{
throw new ArgumentException(
$"Number of voice styles ({voiceStylePaths.Count}) must match number of texts ({textList.Count})");
}
int bsz = voiceStylePaths.Count;
// --- 2. Load Text to Speech --- //
var textToSpeech = Helper.LoadTextToSpeech(parsedArgs.OnnxDir, parsedArgs.UseGpu);
Console.WriteLine();
// --- 3. Load Voice Style --- //
var style = Helper.LoadVoiceStyle(voiceStylePaths, verbose: true);
// --- 4. Synthesize speech --- //
for (int n = 0; n < nTest; n++)
{
Console.WriteLine($"\n[{n + 1}/{nTest}] Starting synthesis...");
var (wav, duration) = Helper.Timer("Generating speech from text", () =>
{
if (batch)
{
return textToSpeech.Batch(textList, style, totalStep, speed);
}
else
{
return textToSpeech.Call(textList[0], style, totalStep, speed);
}
});
if (!Directory.Exists(saveDir))
{
Directory.CreateDirectory(saveDir);
}
for (int b = 0; b < bsz; b++)
{
string fname = $"{Helper.SanitizeFilename(textList[b], 20)}_{n + 1}.wav";
int wavLen = (int)(textToSpeech.SampleRate * duration[b]);
var wavOut = new float[wavLen];
Array.Copy(wav, b * wav.Length / bsz, wavOut, 0, Math.Min(wavLen, wav.Length / bsz));
string outputPath = Path.Combine(saveDir, fname);
Helper.WriteWavFile(outputPath, wavOut, textToSpeech.SampleRate);
Console.WriteLine($"Saved: {outputPath}");
}
}
Console.WriteLine("\n=== Synthesis completed successfully! ===");
}
}
}

881
Cs_HMI/SubProject/SuperTonic/Helper.cs Normal file
View File

@@ -0,0 +1,881 @@
using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Text;
using System.Text.Json;
using System.Text.RegularExpressions;
using Microsoft.ML.OnnxRuntime;
using Microsoft.ML.OnnxRuntime.Tensors;
namespace Supertonic
{
// ============================================================================
// Configuration classes
// ============================================================================
public class Config
{
public AEConfig AE { get; set; } = null;
public TTLConfig TTL { get; set; } = null;
public class AEConfig
{
public int SampleRate { get; set; }
public int BaseChunkSize { get; set; }
}
public class TTLConfig
{
public int ChunkCompressFactor { get; set; }
public int LatentDim { get; set; }
}
}
// ============================================================================
// Style class
// ============================================================================
public class Style
{
public float[] Ttl { get; set; }
public long[] TtlShape { get; set; }
public float[] Dp { get; set; }
public long[] DpShape { get; set; }
public Style(float[] ttl, long[] ttlShape, float[] dp, long[] dpShape)
{
Ttl = ttl;
TtlShape = ttlShape;
Dp = dp;
DpShape = dpShape;
}
}
// ============================================================================
// Unicode text processor
// ============================================================================
public class UnicodeProcessor
{
private readonly Dictionary<int, long> _indexer;
public UnicodeProcessor(string unicodeIndexerPath)
{
var json = File.ReadAllText(unicodeIndexerPath);
var indexerArray = JsonSerializer.Deserialize<long[]>(json) ?? throw new Exception("Failed to load indexer");
_indexer = new Dictionary<int, long>();
for (int i = 0; i < indexerArray.Length; i++)
{
_indexer[i] = indexerArray[i];
}
}
private static string RemoveEmojis(string text)
{
var result = new StringBuilder();
for (int i = 0; i < text.Length; i++)
{
int codePoint;
if (char.IsHighSurrogate(text[i]) && i + 1 < text.Length && char.IsLowSurrogate(text[i + 1]))
{
// Get the full code point from surrogate pair
codePoint = char.ConvertToUtf32(text[i], text[i + 1]);
i++; // Skip the low surrogate
}
else
{
codePoint = text[i];
}
// Check if code point is in emoji ranges
bool isEmoji = (codePoint >= 0x1F600 && codePoint <= 0x1F64F) ||
(codePoint >= 0x1F300 && codePoint <= 0x1F5FF) ||
(codePoint >= 0x1F680 && codePoint <= 0x1F6FF) ||
(codePoint >= 0x1F700 && codePoint <= 0x1F77F) ||
(codePoint >= 0x1F780 && codePoint <= 0x1F7FF) ||
(codePoint >= 0x1F800 && codePoint <= 0x1F8FF) ||
(codePoint >= 0x1F900 && codePoint <= 0x1F9FF) ||
(codePoint >= 0x1FA00 && codePoint <= 0x1FA6F) ||
(codePoint >= 0x1FA70 && codePoint <= 0x1FAFF) ||
(codePoint >= 0x2600 && codePoint <= 0x26FF) ||
(codePoint >= 0x2700 && codePoint <= 0x27BF) ||
(codePoint >= 0x1F1E6 && codePoint <= 0x1F1FF);
if (!isEmoji)
{
if (codePoint > 0xFFFF)
{
// Add back as surrogate pair
result.Append(char.ConvertFromUtf32(codePoint));
}
else
{
result.Append((char)codePoint);
}
}
}
return result.ToString();
}
private string PreprocessText(string text)
{
// TODO: Need advanced normalizer for better performance
text = text.Normalize(NormalizationForm.FormKD);
// FIXME: this should be fixed for non-English languages
// Remove emojis (wide Unicode range)
// C# doesn't support \u{...} syntax in regex, so we use character filtering instead
text = RemoveEmojis(text);
// Replace various dashes and symbols
var replacements = new Dictionary<string, string>
{
{"–", "-"}, // en dash
{"‑", "-"}, // non-breaking hyphen
{"—", "-"}, // em dash
{"¯", " "}, // macron
{"_", " "}, // underscore
{"\u201C", "\""}, // left double quote
{"\u201D", "\""}, // right double quote
{"\u2018", "'"}, // left single quote
{"\u2019", "'"}, // right single quote
{"´", "'"}, // acute accent
{"`", "'"}, // grave accent
{"[", " "}, // left bracket
{"]", " "}, // right bracket
{"|", " "}, // vertical bar
{"/", " "}, // slash
{"#", " "}, // hash
{"→", " "}, // right arrow
{"←", " "}, // left arrow
};
foreach (var kvp in replacements)
{
text = text.Replace(kvp.Key, kvp.Value);
}
// Remove combining diacritics // FIXME: this should be fixed for non-English languages
text = Regex.Replace(text, @"[\u0302\u0303\u0304\u0305\u0306\u0307\u0308\u030A\u030B\u030C\u0327\u0328\u0329\u032A\u032B\u032C\u032D\u032E\u032F]", "");
// Remove special symbols
text = Regex.Replace(text, @"[♥☆♡©\\]", "");
// Replace known expressions
var exprReplacements = new Dictionary<string, string>
{
{"@", " at "},
{"e.g.,", "for example, "},
{"i.e.,", "that is, "},
};
foreach (var kvp in exprReplacements)
{
text = text.Replace(kvp.Key, kvp.Value);
}
// Fix spacing around punctuation
text = Regex.Replace(text, @" ,", ",");
text = Regex.Replace(text, @" \.", ".");
text = Regex.Replace(text, @" !", "!");
text = Regex.Replace(text, @" \?", "?");
text = Regex.Replace(text, @" ;", ";");
text = Regex.Replace(text, @" :", ":");
text = Regex.Replace(text, @" '", "'");
// Remove duplicate quotes
while (text.Contains("\"\""))
{
text = text.Replace("\"\"", "\"");
}
while (text.Contains("''"))
{
text = text.Replace("''", "'");
}
while (text.Contains("``"))
{
text = text.Replace("``", "`");
}
// Remove extra spaces
text = Regex.Replace(text, @"\s+", " ").Trim();
// If text doesn't end with punctuation, quotes, or closing brackets, add a period
if (!Regex.IsMatch(text, @"[.!?;:,'\u0022\u201C\u201D\u2018\u2019)\]}…。」』】〉》›»]$"))
{
text += ".";
}
return text;
}
private int[] TextToUnicodeValues(string text)
{
return text.Select(c => (int)c).ToArray();
}
private float[][][] GetTextMask(long[] textIdsLengths)
{
return Helper.LengthToMask(textIdsLengths);
}
public (long[][] textIds, float[][][] textMask) Call(List<string> textList)
{
var processedTexts = textList.Select(t => PreprocessText(t)).ToList();
var textIdsLengths = processedTexts.Select(t => (long)t.Length).ToArray();
long maxLen = textIdsLengths.Max();
var textIds = new long[textList.Count][];
for (int i = 0; i < processedTexts.Count; i++)
{
textIds[i] = new long[maxLen];
var unicodeVals = TextToUnicodeValues(processedTexts[i]);
for (int j = 0; j < unicodeVals.Length; j++)
{
if (_indexer.TryGetValue(unicodeVals[j], out long val))
{
textIds[i][j] = val;
}
}
}
var textMask = GetTextMask(textIdsLengths);
return (textIds, textMask);
}
}
// ============================================================================
// TextToSpeech class
// ============================================================================
public class TextToSpeech
{
private readonly Config _cfgs;
private readonly UnicodeProcessor _textProcessor;
private readonly InferenceSession _dpOrt;
private readonly InferenceSession _textEncOrt;
private readonly InferenceSession _vectorEstOrt;
private readonly InferenceSession _vocoderOrt;
public readonly int SampleRate;
private readonly int _baseChunkSize;
private readonly int _chunkCompressFactor;
private readonly int _ldim;
public TextToSpeech(
Config cfgs,
UnicodeProcessor textProcessor,
InferenceSession dpOrt,
InferenceSession textEncOrt,
InferenceSession vectorEstOrt,
InferenceSession vocoderOrt)
{
_cfgs = cfgs;
_textProcessor = textProcessor;
_dpOrt = dpOrt;
_textEncOrt = textEncOrt;
_vectorEstOrt = vectorEstOrt;
_vocoderOrt = vocoderOrt;
SampleRate = cfgs.AE.SampleRate;
_baseChunkSize = cfgs.AE.BaseChunkSize;
_chunkCompressFactor = cfgs.TTL.ChunkCompressFactor;
_ldim = cfgs.TTL.LatentDim;
}
private (float[][][] noisyLatent, float[][][] latentMask) SampleNoisyLatent(float[] duration)
{
int bsz = duration.Length;
float wavLenMax = duration.Max() * SampleRate;
var wavLengths = duration.Select(d => (long)(d * SampleRate)).ToArray();
int chunkSize = _baseChunkSize * _chunkCompressFactor;
int latentLen = (int)((wavLenMax + chunkSize - 1) / chunkSize);
int latentDim = _ldim * _chunkCompressFactor;
// Generate random noise
var random = new Random();
var noisyLatent = new float[bsz][][];
for (int b = 0; b < bsz; b++)
{
noisyLatent[b] = new float[latentDim][];
for (int d = 0; d < latentDim; d++)
{
noisyLatent[b][d] = new float[latentLen];
for (int t = 0; t < latentLen; t++)
{
// Box-Muller transform for normal distribution
double u1 = 1.0 - random.NextDouble();
double u2 = 1.0 - random.NextDouble();
noisyLatent[b][d][t] = (float)(Math.Sqrt(-2.0 * Math.Log(u1)) * Math.Cos(2.0 * Math.PI * u2));
}
}
}
var latentMask = Helper.GetLatentMask(wavLengths, _baseChunkSize, _chunkCompressFactor);
// Apply mask
for (int b = 0; b < bsz; b++)
{
for (int d = 0; d < latentDim; d++)
{
for (int t = 0; t < latentLen; t++)
{
noisyLatent[b][d][t] *= latentMask[b][0][t];
}
}
}
return (noisyLatent, latentMask);
}
private (float[] wav, float[] duration) _Infer(List<string> textList, Style style, int totalStep, float speed = 1.05f)
{
int bsz = textList.Count;
if (bsz != style.TtlShape[0])
{
throw new ArgumentException("Number of texts must match number of style vectors");
}
// Process text
var (textIds, textMask) = _textProcessor.Call(textList);
var textIdsShape = new long[] { bsz, textIds[0].Length };
var textMaskShape = new long[] { bsz, 1, textMask[0][0].Length };
var textIdsTensor = Helper.IntArrayToTensor(textIds, textIdsShape);
var textMaskTensor = Helper.ArrayToTensor(textMask, textMaskShape);
var styleTtlTensor = new DenseTensor<float>(style.Ttl, style.TtlShape.Select(x => (int)x).ToArray());
var styleDpTensor = new DenseTensor<float>(style.Dp, style.DpShape.Select(x => (int)x).ToArray());
// Run duration predictor
var dpInputs = new List<NamedOnnxValue>
{
NamedOnnxValue.CreateFromTensor("text_ids", textIdsTensor),
NamedOnnxValue.CreateFromTensor("style_dp", styleDpTensor),
NamedOnnxValue.CreateFromTensor("text_mask", textMaskTensor)
};
using (var dpOutputs = _dpOrt.Run(dpInputs))
{
var durOnnx = dpOutputs.First(o => o.Name == "duration").AsTensor<float>().ToArray();
// Apply speed factor to duration
for (int i = 0; i < durOnnx.Length; i++)
{
durOnnx[i] /= speed;
}
// Run text encoder
var textEncInputs = new List<NamedOnnxValue>
{
NamedOnnxValue.CreateFromTensor("text_ids", textIdsTensor),
NamedOnnxValue.CreateFromTensor("style_ttl", styleTtlTensor),
NamedOnnxValue.CreateFromTensor("text_mask", textMaskTensor)
};
using (var textEncOutputs = _textEncOrt.Run(textEncInputs))
{
var textEmbTensor = textEncOutputs.First(o => o.Name == "text_emb").AsTensor<float>();
// Sample noisy latent
var (xt, latentMask) = SampleNoisyLatent(durOnnx);
var latentShape = new long[] { bsz, xt[0].Length, xt[0][0].Length };
var latentMaskShape = new long[] { bsz, 1, latentMask[0][0].Length };
var totalStepArray = Enumerable.Repeat((float)totalStep, bsz).ToArray();
// Iterative denoising
for (int step = 0; step < totalStep; step++)
{
var currentStepArray = Enumerable.Repeat((float)step, bsz).ToArray();
var vectorEstInputs = new List<NamedOnnxValue>
{
NamedOnnxValue.CreateFromTensor("noisy_latent", Helper.ArrayToTensor(xt, latentShape)),
NamedOnnxValue.CreateFromTensor("text_emb", textEmbTensor),
NamedOnnxValue.CreateFromTensor("style_ttl", styleTtlTensor),
NamedOnnxValue.CreateFromTensor("text_mask", textMaskTensor),
NamedOnnxValue.CreateFromTensor("latent_mask", Helper.ArrayToTensor(latentMask, latentMaskShape)),
NamedOnnxValue.CreateFromTensor("total_step", new DenseTensor<float>(totalStepArray, new int[] { bsz })),
NamedOnnxValue.CreateFromTensor("current_step", new DenseTensor<float>(currentStepArray, new int[] { bsz }))
};
using (var vectorEstOutputs = _vectorEstOrt.Run(vectorEstInputs))
{
var denoisedLatent = vectorEstOutputs.First(o => o.Name == "denoised_latent").AsTensor<float>();
// Update xt
int idx = 0;
for (int b = 0; b < bsz; b++)
{
for (int d = 0; d < xt[b].Length; d++)
{
for (int t = 0; t < xt[b][d].Length; t++)
{
xt[b][d][t] = denoisedLatent.GetValue(idx++);
}
}
}
}
}
// Run vocoder
var vocoderInputs = new List<NamedOnnxValue>
{
NamedOnnxValue.CreateFromTensor("latent", Helper.ArrayToTensor(xt, latentShape))
};
using (var vocoderOutputs = _vocoderOrt.Run(vocoderInputs))
{
var wavTensor = vocoderOutputs.First(o => o.Name == "wav_tts").AsTensor<float>();
return (wavTensor.ToArray(), durOnnx);
}
}
}
}
public (float[] wav, float[] duration) Call(string text, Style style, int totalStep, float speed = 1.05f, float silenceDuration = 0.3f)
{
if (style.TtlShape[0] != 1)
{
throw new ArgumentException("Single speaker text to speech only supports single style");
}
var textList = Helper.ChunkText(text);
var wavCat = new List<float>();
float durCat = 0.0f;
foreach (var chunk in textList)
{
var (wav, duration) = _Infer(new List<string> { chunk }, style, totalStep, speed);
if (wavCat.Count == 0)
{
wavCat.AddRange(wav);
durCat = duration[0];
}
else
{
int silenceLen = (int)(silenceDuration * SampleRate);
var silence = new float[silenceLen];
wavCat.AddRange(silence);
wavCat.AddRange(wav);
durCat += duration[0] + silenceDuration;
}
}
return (wavCat.ToArray(), new float[] { durCat });
}
public (float[] wav, float[] duration) Batch(List<string> textList, Style style, int totalStep, float speed = 1.05f)
{
return _Infer(textList, style, totalStep, speed);
}
}
// ============================================================================
// Helper class with utility functions
// ============================================================================
public static class Helper
{
// ============================================================================
// Utility functions
// ============================================================================
public static float[][][] LengthToMask(long[] lengths, long maxLen = -1)
{
if (maxLen == -1)
{
maxLen = lengths.Max();
}
var mask = new float[lengths.Length][][];
for (int i = 0; i < lengths.Length; i++)
{
mask[i] = new float[1][];
mask[i][0] = new float[maxLen];
for (int j = 0; j < maxLen; j++)
{
mask[i][0][j] = j < lengths[i] ? 1.0f : 0.0f;
}
}
return mask;
}
public static float[][][] GetLatentMask(long[] wavLengths, int baseChunkSize, int chunkCompressFactor)
{
int latentSize = baseChunkSize * chunkCompressFactor;
var latentLengths = wavLengths.Select(len => (len + latentSize - 1) / latentSize).ToArray();
return LengthToMask(latentLengths);
}
// ============================================================================
// ONNX model loading
// ============================================================================
public static InferenceSession LoadOnnx(string onnxPath, SessionOptions opts)
{
return new InferenceSession(onnxPath, opts);
}
public static (InferenceSession dp, InferenceSession textEnc, InferenceSession vectorEst, InferenceSession vocoder)
LoadOnnxAll(string onnxDir, SessionOptions opts)
{
var dpPath = Path.Combine(onnxDir, "duration_predictor.onnx");
var textEncPath = Path.Combine(onnxDir, "text_encoder.onnx");
var vectorEstPath = Path.Combine(onnxDir, "vector_estimator.onnx");
var vocoderPath = Path.Combine(onnxDir, "vocoder.onnx");
return (
LoadOnnx(dpPath, opts),
LoadOnnx(textEncPath, opts),
LoadOnnx(vectorEstPath, opts),
LoadOnnx(vocoderPath, opts)
);
}
// ============================================================================
// Configuration loading
// ============================================================================
public static Config LoadCfgs(string onnxDir)
{
var cfgPath = Path.Combine(onnxDir, "tts.json");
var json = File.ReadAllText(cfgPath);
using (var doc = JsonDocument.Parse(json))
{
var root = doc.RootElement;
return new Config
{
AE = new Config.AEConfig
{
SampleRate = root.GetProperty("ae").GetProperty("sample_rate").GetInt32(),
BaseChunkSize = root.GetProperty("ae").GetProperty("base_chunk_size").GetInt32()
},
TTL = new Config.TTLConfig
{
ChunkCompressFactor = root.GetProperty("ttl").GetProperty("chunk_compress_factor").GetInt32(),
LatentDim = root.GetProperty("ttl").GetProperty("latent_dim").GetInt32()
}
};
}
}
public static UnicodeProcessor LoadTextProcessor(string onnxDir)
{
var unicodeIndexerPath = Path.Combine(onnxDir, "unicode_indexer.json");
return new UnicodeProcessor(unicodeIndexerPath);
}
// ============================================================================
// Voice style loading
// ============================================================================
public static Style LoadVoiceStyle(List<string> voiceStylePaths, bool verbose = false)
{
int bsz = voiceStylePaths.Count;
// Read first file to get dimensions
var firstJson = File.ReadAllText(voiceStylePaths[0]);
using (var firstDoc = JsonDocument.Parse(firstJson))
{
var firstRoot = firstDoc.RootElement;
var ttlDims = ParseInt64Array(firstRoot.GetProperty("style_ttl").GetProperty("dims"));
var dpDims = ParseInt64Array(firstRoot.GetProperty("style_dp").GetProperty("dims"));
long ttlDim1 = ttlDims[1];
long ttlDim2 = ttlDims[2];
long dpDim1 = dpDims[1];
long dpDim2 = dpDims[2];
// Pre-allocate arrays with full batch size
int ttlSize = (int)(bsz * ttlDim1 * ttlDim2);
int dpSize = (int)(bsz * dpDim1 * dpDim2);
var ttlFlat = new float[ttlSize];
var dpFlat = new float[dpSize];
// Fill in the data
for (int i = 0; i < bsz; i++)
{
var json = File.ReadAllText(voiceStylePaths[i]);
using (var doc = JsonDocument.Parse(json))
{
var root = doc.RootElement;
// Flatten data
var ttlData3D = ParseFloat3DArray(root.GetProperty("style_ttl").GetProperty("data"));
var ttlDataFlat = new List<float>();
foreach (var batch in ttlData3D)
{
foreach (var row in batch)
{
ttlDataFlat.AddRange(row);
}
}
var dpData3D = ParseFloat3DArray(root.GetProperty("style_dp").GetProperty("data"));
var dpDataFlat = new List<float>();
foreach (var batch in dpData3D)
{
foreach (var row in batch)
{
dpDataFlat.AddRange(row);
}
}
// Copy to pre-allocated array
int ttlOffset = (int)(i * ttlDim1 * ttlDim2);
ttlDataFlat.CopyTo(ttlFlat, ttlOffset);
int dpOffset = (int)(i * dpDim1 * dpDim2);
dpDataFlat.CopyTo(dpFlat, dpOffset);
}
}
var ttlShape = new long[] { bsz, ttlDim1, ttlDim2 };
var dpShape = new long[] { bsz, dpDim1, dpDim2 };
if (verbose)
{
Console.WriteLine($"Loaded {bsz} voice styles");
}
return new Style(ttlFlat, ttlShape, dpFlat, dpShape);
}
}
private static float[][][] ParseFloat3DArray(JsonElement element)
{
var result = new List<float[][]>();
foreach (var batch in element.EnumerateArray())
{
var batch2D = new List<float[]>();
foreach (var row in batch.EnumerateArray())
{
var rowData = new List<float>();
foreach (var val in row.EnumerateArray())
{
rowData.Add(val.GetSingle());
}
batch2D.Add(rowData.ToArray());
}
result.Add(batch2D.ToArray());
}
return result.ToArray();
}
private static long[] ParseInt64Array(JsonElement element)
{
var result = new List<long>();
foreach (var val in element.EnumerateArray())
{
result.Add(val.GetInt64());
}
return result.ToArray();
}
// ============================================================================
// TextToSpeech loading
// ============================================================================
public static TextToSpeech LoadTextToSpeech(string onnxDir, bool useGpu = false)
{
var opts = new SessionOptions();
if (useGpu)
{
throw new NotImplementedException("GPU mode is not supported yet");
}
else
{
Console.WriteLine("Using CPU for inference");
}
var cfgs = LoadCfgs(onnxDir);
var (dpOrt, textEncOrt, vectorEstOrt, vocoderOrt) = LoadOnnxAll(onnxDir, opts);
var textProcessor = LoadTextProcessor(onnxDir);
return new TextToSpeech(cfgs, textProcessor, dpOrt, textEncOrt, vectorEstOrt, vocoderOrt);
}
// ============================================================================
// WAV file writing
// ============================================================================
public static void WriteWavFile(string filename, float[] audioData, int sampleRate)
{
using (var writer = new BinaryWriter(File.Open(filename, FileMode.Create)))
{
int numChannels = 1;
int bitsPerSample = 16;
int byteRate = sampleRate * numChannels * bitsPerSample / 8;
short blockAlign = (short)(numChannels * bitsPerSample / 8);
int dataSize = audioData.Length * bitsPerSample / 8;
// RIFF header
writer.Write(Encoding.ASCII.GetBytes("RIFF"));
writer.Write(36 + dataSize);
writer.Write(Encoding.ASCII.GetBytes("WAVE"));
// fmt chunk
writer.Write(Encoding.ASCII.GetBytes("fmt "));
writer.Write(16); // fmt chunk size
writer.Write((short)1); // audio format (PCM)
writer.Write((short)numChannels);
writer.Write(sampleRate);
writer.Write(byteRate);
writer.Write(blockAlign);
writer.Write((short)bitsPerSample);
// data chunk
writer.Write(Encoding.ASCII.GetBytes("data"));
writer.Write(dataSize);
// Write audio data
foreach (var sample in audioData)
{
float clamped = Math.Max(-1.0f, Math.Min(1.0f, sample));
short intSample = (short)(clamped * 32767);
writer.Write(intSample);
}
}
}
// ============================================================================
// Tensor conversion utilities
// ============================================================================
public static DenseTensor<float> ArrayToTensor(float[][][] array, long[] dims)
{
var flat = new List<float>();
foreach (var batch in array)
{
foreach (var row in batch)
{
flat.AddRange(row);
}
}
return new DenseTensor<float>(flat.ToArray(), dims.Select(x => (int)x).ToArray());
}
public static DenseTensor<long> IntArrayToTensor(long[][] array, long[] dims)
{
var flat = new List<long>();
foreach (var row in array)
{
flat.AddRange(row);
}
return new DenseTensor<long>(flat.ToArray(), dims.Select(x => (int)x).ToArray());
}
// ============================================================================
// Timer utility
// ============================================================================
public static T Timer<T>(string name, Func<T> func)
{
var start = DateTime.Now;
Console.WriteLine($"{name}...");
var result = func();
var elapsed = (DateTime.Now - start).TotalSeconds;
Console.WriteLine($" -> {name} completed in {elapsed:F2} sec");
return result;
}
public static string SanitizeFilename(string text, int maxLen)
{
var result = new StringBuilder();
int count = 0;
foreach (char c in text)
{
if (count >= maxLen) break;
if (char.IsLetterOrDigit(c))
{
result.Append(c);
}
else
{
result.Append('_');
}
count++;
}
return result.ToString();
}
// ============================================================================
// Chunk text
// ============================================================================
public static List<string> ChunkText(string text, int maxLen = 300)
{
var chunks = new List<string>();
// Split by paragraph (two or more newlines)
var paragraphRegex = new Regex(@"\n\s*\n+");
var paragraphs = paragraphRegex.Split(text.Trim())
.Select(p => p.Trim())
.Where(p => !string.IsNullOrEmpty(p))
.ToList();
// Split by sentence boundaries, excluding abbreviations
var sentenceRegex = new Regex(@"(?<!Mr\.|Mrs\.|Ms\.|Dr\.|Prof\.|Sr\.|Jr\.|Ph\.D\.|etc\.|e\.g\.|i\.e\.|vs\.|Inc\.|Ltd\.|Co\.|Corp\.|St\.|Ave\.|Blvd\.)(?<!\b[A-Z]\.)(?<=[.!?])\s+");
foreach (var paragraph in paragraphs)
{
var sentences = sentenceRegex.Split(paragraph);
string currentChunk = "";
foreach (var sentence in sentences)
{
if (string.IsNullOrEmpty(sentence)) continue;
if (currentChunk.Length + sentence.Length + 1 <= maxLen)
{
if (!string.IsNullOrEmpty(currentChunk))
{
currentChunk += " ";
}
currentChunk += sentence;
}
else
{
if (!string.IsNullOrEmpty(currentChunk))
{
chunks.Add(currentChunk.Trim());
}
currentChunk = sentence;
}
}
if (!string.IsNullOrEmpty(currentChunk))
{
chunks.Add(currentChunk.Trim());
}
}
// If no chunks were created, return the original text
if (chunks.Count == 0)
{
chunks.Add(text.Trim());
}
return chunks;
}
}
}

36
Cs_HMI/SubProject/SuperTonic/Properties/AssemblyInfo.cs Normal file
View File

@@ -0,0 +1,36 @@
using System.Reflection;
using System.Runtime.CompilerServices;
using System.Runtime.InteropServices;
// 어셈블리에 대한 일반 정보는 다음 특성 집합을 통해
// 제어됩니다. 어셈블리와 관련된 정보를 수정하려면
// 이러한 특성 값을 변경하세요.
[assembly: AssemblyTitle("ClassLibrary1")]
[assembly: AssemblyDescription("")]
[assembly: AssemblyConfiguration("")]
[assembly: AssemblyCompany("")]
[assembly: AssemblyProduct("ClassLibrary1")]
[assembly: AssemblyCopyright("Copyright © 2025")]
[assembly: AssemblyTrademark("")]
[assembly: AssemblyCulture("")]
// ComVisible을 false로 설정하면 이 어셈블리의 형식이 COM 구성 요소에
// 표시되지 않습니다. COM에서 이 어셈블리의 형식에 액세스하려면
// 해당 형식에 대해 ComVisible 특성을 true로 설정하세요.
[assembly: ComVisible(false)]
// 이 프로젝트가 COM에 노출되는 경우 다음 GUID는 typelib의 ID를 나타냅니다.
[assembly: Guid("19675e19-eb91-493e-88c3-32b3c094b749")]
// 어셈블리의 버전 정보는 다음 네 가지 값으로 구성됩니다.
//
// 주 버전
// 부 버전
// 빌드 번호
// 수정 버전
//
// 모든 값을 지정하거나 아래와 같이 '*'를 사용하여 빌드 번호 및 수정 번호를
// 기본값으로 할 수 있습니다.
// [assembly: AssemblyVersion("1.0.*")]
[assembly: AssemblyVersion("1.0.0.0")]
[assembly: AssemblyFileVersion("1.0.0.0")]

126
Cs_HMI/SubProject/SuperTonic/Supertonic.Netfx48.csproj Normal file
View File

@@ -0,0 +1,126 @@
<?xml version="1.0" encoding="utf-8"?>
<Project ToolsVersion="15.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
<Import Project="..\csharp\packages\Microsoft.ML.OnnxRuntime.1.23.2\build\netstandard2.0\Microsoft.ML.OnnxRuntime.props" Condition="Exists('..\csharp\packages\Microsoft.ML.OnnxRuntime.1.23.2\build\netstandard2.0\Microsoft.ML.OnnxRuntime.props')" />
<Import Project="$(MSBuildExtensionsPath)\$(MSBuildToolsVersion)\Microsoft.Common.props" Condition="Exists('$(MSBuildExtensionsPath)\$(MSBuildToolsVersion)\Microsoft.Common.props')" />
<PropertyGroup>
<Configuration Condition=" '$(Configuration)' == '' ">Debug</Configuration>
<Platform Condition=" '$(Platform)' == '' ">AnyCPU</Platform>
<ProjectGuid>{19675E19-EB91-493E-88C3-32B3C094B749}</ProjectGuid>
<OutputType>Exe</OutputType>
<AppDesignerFolder>Properties</AppDesignerFolder>
<RootNamespace>Supertonic</RootNamespace>
<AssemblyName>Supertonic.Net48</AssemblyName>
<TargetFrameworkVersion>v4.8</TargetFrameworkVersion>
<FileAlignment>512</FileAlignment>
<Deterministic>true</Deterministic>
<NuGetPackageImportStamp>
</NuGetPackageImportStamp>
</PropertyGroup>
<PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Debug|AnyCPU' ">
<DebugSymbols>true</DebugSymbols>
<DebugType>full</DebugType>
<Optimize>false</Optimize>
<OutputPath>bin\Debug\</OutputPath>
<DefineConstants>DEBUG;TRACE</DefineConstants>
<ErrorReport>prompt</ErrorReport>
<WarningLevel>4</WarningLevel>
</PropertyGroup>
<PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Release|AnyCPU' ">
<DebugType>pdbonly</DebugType>
<Optimize>true</Optimize>
<OutputPath>bin\Release\</OutputPath>
<DefineConstants>TRACE</DefineConstants>
<ErrorReport>prompt</ErrorReport>
<WarningLevel>4</WarningLevel>
</PropertyGroup>
<PropertyGroup>
<StartupObject>Supertonic.Program</StartupObject>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)|$(Platform)' == 'Debug|Win32'">
<PlatformTarget>x64</PlatformTarget>
<OutputPath>bin\Debug\</OutputPath>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)|$(Platform)' == 'Debug|x64'">
<DebugSymbols>true</DebugSymbols>
<OutputPath>bin\x64\Debug\</OutputPath>
<DefineConstants>DEBUG;TRACE</DefineConstants>
<DebugType>full</DebugType>
<PlatformTarget>x64</PlatformTarget>
<ErrorReport>prompt</ErrorReport>
<CodeAnalysisRuleSet>MinimumRecommendedRules.ruleset</CodeAnalysisRuleSet>
<Prefer32Bit>true</Prefer32Bit>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)|$(Platform)' == 'Release|x64'">
<OutputPath>bin\x64\Release\</OutputPath>
<DefineConstants>TRACE</DefineConstants>
<Optimize>true</Optimize>
<DebugType>pdbonly</DebugType>
<PlatformTarget>x64</PlatformTarget>
<ErrorReport>prompt</ErrorReport>
<CodeAnalysisRuleSet>MinimumRecommendedRules.ruleset</CodeAnalysisRuleSet>
<Prefer32Bit>true</Prefer32Bit>
</PropertyGroup>
<ItemGroup>
<Reference Include="Microsoft.Bcl.AsyncInterfaces, Version=10.0.0.1, Culture=neutral, PublicKeyToken=cc7b13ffcd2ddd51, processorArchitecture=MSIL">
<HintPath>..\csharp\packages\Microsoft.Bcl.AsyncInterfaces.10.0.1\lib\net462\Microsoft.Bcl.AsyncInterfaces.dll</HintPath>
</Reference>
<Reference Include="Microsoft.ML.OnnxRuntime, Version=1.23.2.0, Culture=neutral, PublicKeyToken=f27f157f0a5b7bb6, processorArchitecture=MSIL">
<HintPath>..\csharp\packages\Microsoft.ML.OnnxRuntime.Managed.1.23.2\lib\netstandard2.0\Microsoft.ML.OnnxRuntime.dll</HintPath>
</Reference>
<Reference Include="System" />
<Reference Include="System.Buffers, Version=4.0.5.0, Culture=neutral, PublicKeyToken=cc7b13ffcd2ddd51, processorArchitecture=MSIL">
<HintPath>..\csharp\packages\System.Buffers.4.6.1\lib\net462\System.Buffers.dll</HintPath>
</Reference>
<Reference Include="System.Core" />
<Reference Include="System.IO.Pipelines, Version=10.0.0.1, Culture=neutral, PublicKeyToken=cc7b13ffcd2ddd51, processorArchitecture=MSIL">
<HintPath>..\csharp\packages\System.IO.Pipelines.10.0.1\lib\net462\System.IO.Pipelines.dll</HintPath>
</Reference>
<Reference Include="System.Memory, Version=4.0.5.0, Culture=neutral, PublicKeyToken=cc7b13ffcd2ddd51, processorArchitecture=MSIL">
<HintPath>..\csharp\packages\System.Memory.4.6.3\lib\net462\System.Memory.dll</HintPath>
</Reference>
<Reference Include="System.Numerics" />
<Reference Include="System.Numerics.Vectors, Version=4.1.6.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a, processorArchitecture=MSIL">
<HintPath>..\csharp\packages\System.Numerics.Vectors.4.6.1\lib\net462\System.Numerics.Vectors.dll</HintPath>
</Reference>
<Reference Include="System.Runtime.CompilerServices.Unsafe, Version=6.0.3.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a, processorArchitecture=MSIL">
<HintPath>..\csharp\packages\System.Runtime.CompilerServices.Unsafe.6.1.2\lib\net462\System.Runtime.CompilerServices.Unsafe.dll</HintPath>
</Reference>
<Reference Include="System.Text.Encodings.Web, Version=10.0.0.1, Culture=neutral, PublicKeyToken=cc7b13ffcd2ddd51, processorArchitecture=MSIL">
<HintPath>..\csharp\packages\System.Text.Encodings.Web.10.0.1\lib\net462\System.Text.Encodings.Web.dll</HintPath>
</Reference>
<Reference Include="System.Text.Json, Version=10.0.0.1, Culture=neutral, PublicKeyToken=cc7b13ffcd2ddd51, processorArchitecture=MSIL">
<HintPath>..\csharp\packages\System.Text.Json.10.0.1\lib\net462\System.Text.Json.dll</HintPath>
</Reference>
<Reference Include="System.Threading.Tasks.Extensions, Version=4.2.4.0, Culture=neutral, PublicKeyToken=cc7b13ffcd2ddd51, processorArchitecture=MSIL">
<HintPath>..\csharp\packages\System.Threading.Tasks.Extensions.4.6.3\lib\net462\System.Threading.Tasks.Extensions.dll</HintPath>
</Reference>
<Reference Include="System.Xml.Linq" />
<Reference Include="System.Data.DataSetExtensions" />
<Reference Include="Microsoft.CSharp" />
<Reference Include="System.Data" />
<Reference Include="System.Net.Http" />
<Reference Include="System.Xml" />
</ItemGroup>
<ItemGroup>
<Compile Include="ExampleONNX.cs" />
<Compile Include="Helper.cs" />
<Compile Include="Properties\AssemblyInfo.cs" />
</ItemGroup>
<ItemGroup>
<None Include="app.config" />
<None Include="packages.config" />
</ItemGroup>
<Import Project="$(MSBuildToolsPath)\Microsoft.CSharp.targets" />
<Import Project="..\csharp\packages\Microsoft.ML.OnnxRuntime.Managed.1.23.2\build\netstandard2.0\Microsoft.ML.OnnxRuntime.Managed.targets" Condition="Exists('..\csharp\packages\Microsoft.ML.OnnxRuntime.Managed.1.23.2\build\netstandard2.0\Microsoft.ML.OnnxRuntime.Managed.targets')" />
<Target Name="EnsureNuGetPackageBuildImports" BeforeTargets="PrepareForBuild">
<PropertyGroup>
<ErrorText>이 프로젝트는 이 컴퓨터에 없는 NuGet 패키지를 참조합니다. 해당 패키지를 다운로드하려면 NuGet 패키지 복원을 사용하십시오. 자세한 내용은 http://go.microsoft.com/fwlink/?LinkID=322105를 참조하십시오. 누락된 파일은 {0}입니다.</ErrorText>
</PropertyGroup>
<Error Condition="!Exists('..\csharp\packages\Microsoft.ML.OnnxRuntime.Managed.1.23.2\build\netstandard2.0\Microsoft.ML.OnnxRuntime.Managed.targets')" Text="$([System.String]::Format('$(ErrorText)', '..\csharp\packages\Microsoft.ML.OnnxRuntime.Managed.1.23.2\build\netstandard2.0\Microsoft.ML.OnnxRuntime.Managed.targets'))" />
<Error Condition="!Exists('..\csharp\packages\Microsoft.ML.OnnxRuntime.1.23.2\build\netstandard2.0\Microsoft.ML.OnnxRuntime.props')" Text="$([System.String]::Format('$(ErrorText)', '..\csharp\packages\Microsoft.ML.OnnxRuntime.1.23.2\build\netstandard2.0\Microsoft.ML.OnnxRuntime.props'))" />
<Error Condition="!Exists('..\csharp\packages\Microsoft.ML.OnnxRuntime.1.23.2\build\netstandard2.0\Microsoft.ML.OnnxRuntime.targets')" Text="$([System.String]::Format('$(ErrorText)', '..\csharp\packages\Microsoft.ML.OnnxRuntime.1.23.2\build\netstandard2.0\Microsoft.ML.OnnxRuntime.targets'))" />
<Error Condition="!Exists('..\csharp\packages\System.ValueTuple.4.6.1\build\net471\System.ValueTuple.targets')" Text="$([System.String]::Format('$(ErrorText)', '..\csharp\packages\System.ValueTuple.4.6.1\build\net471\System.ValueTuple.targets'))" />
</Target>
<Import Project="..\csharp\packages\Microsoft.ML.OnnxRuntime.1.23.2\build\netstandard2.0\Microsoft.ML.OnnxRuntime.targets" Condition="Exists('..\csharp\packages\Microsoft.ML.OnnxRuntime.1.23.2\build\netstandard2.0\Microsoft.ML.OnnxRuntime.targets')" />
<Import Project="..\csharp\packages\System.ValueTuple.4.6.1\build\net471\System.ValueTuple.targets" Condition="Exists('..\csharp\packages\System.ValueTuple.4.6.1\build\net471\System.ValueTuple.targets')" />
</Project>

49
Cs_HMI/SubProject/SuperTonic/Supertonic.Netfx48.sln Normal file
View File

@@ -0,0 +1,49 @@

Microsoft Visual Studio Solution File, Format Version 12.00
# Visual Studio Express 15 for Windows Desktop
VisualStudioVersion = 15.0.36324.19
MinimumVisualStudioVersion = 10.0.40219.1
Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "Supertonic.Netfx48", "Supertonic.Netfx48.csproj", "{19675E19-EB91-493E-88C3-32B3C094B749}"
EndProject
Global
GlobalSection(SolutionConfigurationPlatforms) = preSolution
Debug|Any CPU = Debug|Any CPU
Debug|ARM = Debug|ARM
Debug|ARM64 = Debug|ARM64
Debug|x64 = Debug|x64
Debug|x86 = Debug|x86
Release|Any CPU = Release|Any CPU
Release|ARM = Release|ARM
Release|ARM64 = Release|ARM64
Release|x64 = Release|x64
Release|x86 = Release|x86
EndGlobalSection
GlobalSection(ProjectConfigurationPlatforms) = postSolution
{19675E19-EB91-493E-88C3-32B3C094B749}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
{19675E19-EB91-493E-88C3-32B3C094B749}.Debug|Any CPU.Build.0 = Debug|Any CPU
{19675E19-EB91-493E-88C3-32B3C094B749}.Debug|ARM.ActiveCfg = Debug|ARM
{19675E19-EB91-493E-88C3-32B3C094B749}.Debug|ARM.Build.0 = Debug|ARM
{19675E19-EB91-493E-88C3-32B3C094B749}.Debug|ARM64.ActiveCfg = Debug|ARM64
{19675E19-EB91-493E-88C3-32B3C094B749}.Debug|ARM64.Build.0 = Debug|ARM64
{19675E19-EB91-493E-88C3-32B3C094B749}.Debug|x64.ActiveCfg = Debug|x64
{19675E19-EB91-493E-88C3-32B3C094B749}.Debug|x64.Build.0 = Debug|x64
{19675E19-EB91-493E-88C3-32B3C094B749}.Debug|x86.ActiveCfg = Debug|Win32
{19675E19-EB91-493E-88C3-32B3C094B749}.Debug|x86.Build.0 = Debug|Win32
{19675E19-EB91-493E-88C3-32B3C094B749}.Release|Any CPU.ActiveCfg = Release|Any CPU
{19675E19-EB91-493E-88C3-32B3C094B749}.Release|Any CPU.Build.0 = Release|Any CPU
{19675E19-EB91-493E-88C3-32B3C094B749}.Release|ARM.ActiveCfg = Release|ARM
{19675E19-EB91-493E-88C3-32B3C094B749}.Release|ARM.Build.0 = Release|ARM
{19675E19-EB91-493E-88C3-32B3C094B749}.Release|ARM64.ActiveCfg = Release|ARM64
{19675E19-EB91-493E-88C3-32B3C094B749}.Release|ARM64.Build.0 = Release|ARM64
{19675E19-EB91-493E-88C3-32B3C094B749}.Release|x64.ActiveCfg = Release|x64
{19675E19-EB91-493E-88C3-32B3C094B749}.Release|x64.Build.0 = Release|x64
{19675E19-EB91-493E-88C3-32B3C094B749}.Release|x86.ActiveCfg = Release|Win32
{19675E19-EB91-493E-88C3-32B3C094B749}.Release|x86.Build.0 = Release|Win32
EndGlobalSection
GlobalSection(SolutionProperties) = preSolution
HideSolutionNode = FALSE
EndGlobalSection
GlobalSection(ExtensibilityGlobals) = postSolution
SolutionGuid = {5F2E20C5-E704-4B99-8FE9-54394113916E}
EndGlobalSection
EndGlobal

11
Cs_HMI/SubProject/SuperTonic/app.config Normal file
View File

@@ -0,0 +1,11 @@
<?xml version="1.0" encoding="utf-8"?>
<configuration>
<runtime>
<assemblyBinding xmlns="urn:schemas-microsoft-com:asm.v1">
<dependentAssembly>
<assemblyIdentity name="System.Memory" publicKeyToken="cc7b13ffcd2ddd51" culture="neutral" />
<bindingRedirect oldVersion="0.0.0.0-4.0.5.0" newVersion="4.0.5.0" />
</dependentAssembly>
</assemblyBinding>
</runtime>
</configuration>

BIN
Cs_HMI/SubProject/SuperTonic/assets/onnx/duration_predictor.onnx Normal file
View File

Binary file not shown.

BIN
Cs_HMI/SubProject/SuperTonic/assets/onnx/text_encoder.onnx Normal file
View File

Binary file not shown.

316
Cs_HMI/SubProject/SuperTonic/assets/onnx/tts.json Normal file
View File

@@ -0,0 +1,316 @@
{
"tts_version": "v1.5.0",
"split": "opensource-en",
"ttl_ckpt_path": "unknown.pt",
"dp_ckpt_path": "unknown.pt",
"ae_ckpt_path": "unknown.pt",
"ttl_train": "unknown",
"dp_train": "unknown",
"ae_train": "unknown",
"ttl": {
"latent_dim": 24,
"chunk_compress_factor": 6,
"batch_expander": {
"n_batch_expand": 6
},
"normalizer": {
"scale": 0.25
},
"text_encoder": {
"char_dict_path": "resources/metadata/char_dict/opensource-en/char_dict.json",
"text_embedder": {
"char_dict_path": "resources/metadata/char_dict/opensource-en/char_dict.json",
"char_emb_dim": 256
},
"convnext": {
"idim": 256,
"ksz": 5,
"intermediate_dim": 1024,
"num_layers": 6,
"dilation_lst": [
1,
1,
1,
1,
1,
1
]
},
"attn_encoder": {
"hidden_channels": 256,
"filter_channels": 1024,
"n_heads": 4,
"n_layers": 4,
"p_dropout": 0.0
},
"proj_out": {
"idim": 256,
"odim": 256
}
},
"flow_matching": {
"sig_min": 0
},
"style_encoder": {
"proj_in": {
"ldim": 24,
"chunk_compress_factor": 6,
"odim": 256
},
"convnext": {
"idim": 256,
"ksz": 5,
"intermediate_dim": 1024,
"num_layers": 6,
"dilation_lst": [
1,
1,
1,
1,
1,
1
]
},
"style_token_layer": {
"input_dim": 256,
"n_style": 50,
"style_key_dim": 256,
"style_value_dim": 256,
"prototype_dim": 256,
"n_units": 256,
"n_heads": 2
}
},
"speech_prompted_text_encoder": {
"text_dim": 256,
"style_dim": 256,
"n_units": 256,
"n_heads": 2
},
"uncond_masker": {
"prob_both_uncond": 0.04,
"prob_text_uncond": 0.01,
"std": 0.1,
"text_dim": 256,
"n_style": 50,
"style_key_dim": 256,
"style_value_dim": 256
},
"vector_field": {
"proj_in": {
"ldim": 24,
"chunk_compress_factor": 6,
"odim": 512
},
"time_encoder": {
"time_dim": 64,
"hdim": 256
},
"main_blocks": {
"n_blocks": 4,
"time_cond_layer": {
"idim": 512,
"time_dim": 64
},
"style_cond_layer": {
"idim": 512,
"style_dim": 256
},
"text_cond_layer": {
"idim": 512,
"text_dim": 256,
"n_heads": 4,
"use_residual": true,
"rotary_base": 10000,
"rotary_scale": 10
},
"convnext_0": {
"idim": 512,
"ksz": 5,
"intermediate_dim": 1024,
"num_layers": 4,
"dilation_lst": [
1,
2,
4,
8
]
},
"convnext_1": {
"idim": 512,
"ksz": 5,
"intermediate_dim": 1024,
"num_layers": 1,
"dilation_lst": [
1
]
},
"convnext_2": {
"idim": 512,
"ksz": 5,
"intermediate_dim": 1024,
"num_layers": 1,
"dilation_lst": [
1
]
}
},
"last_convnext": {
"idim": 512,
"ksz": 5,
"intermediate_dim": 1024,
"num_layers": 4,
"dilation_lst": [
1,
1,
1,
1
]
},
"proj_out": {
"idim": 512,
"chunk_compress_factor": 6,
"ldim": 24
}
}
},
"ae": {
"sample_rate": 44100,
"n_delay": 0,
"base_chunk_size": 512,
"chunk_compress_factor": 1,
"ldim": 24,
"encoder": {
"spec_processor": {
"n_fft": 2048,
"win_length": 2048,
"hop_length": 512,
"n_mels": 228,
"sample_rate": 44100,
"eps": 1e-05,
"norm_mean": 0.0,
"norm_std": 1.0
},
"ksz_init": 7,
"ksz": 7,
"num_layers": 10,
"dilation_lst": [
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
],
"intermediate_dim": 2048,
"idim": 1253,
"hdim": 512,
"odim": 24
},
"decoder": {
"ksz_init": 7,
"ksz": 7,
"num_layers": 10,
"dilation_lst": [
1,
2,
4,
1,
2,
4,
1,
1,
1,
1
],
"intermediate_dim": 2048,
"idim": 24,
"hdim": 512,
"head": {
"idim": 512,
"hdim": 2048,
"odim": 512,
"ksz": 3
}
}
},
"dp": {
"latent_dim": 24,
"chunk_compress_factor": 6,
"normalizer": {
"scale": 1.0
},
"sentence_encoder": {
"char_emb_dim": 64,
"char_dict_path": "resources/metadata/char_dict/opensource-en/char_dict.json",
"text_embedder": {
"char_dict_path": "resources/metadata/char_dict/opensource-en/char_dict.json",
"char_emb_dim": 64
},
"convnext": {
"idim": 64,
"ksz": 5,
"intermediate_dim": 256,
"num_layers": 6,
"dilation_lst": [
1,
1,
1,
1,
1,
1
]
},
"attn_encoder": {
"hidden_channels": 64,
"filter_channels": 256,
"n_heads": 2,
"n_layers": 2,
"p_dropout": 0.0
},
"proj_out": {
"idim": 64,
"odim": 64
}
},
"style_encoder": {
"proj_in": {
"ldim": 24,
"chunk_compress_factor": 6,
"odim": 64
},
"convnext": {
"idim": 64,
"ksz": 5,
"intermediate_dim": 256,
"num_layers": 4,
"dilation_lst": [
1,
1,
1,
1
]
},
"style_token_layer": {
"input_dim": 64,
"n_style": 8,
"style_key_dim": 0,
"style_value_dim": 16,
"prototype_dim": 64,
"n_units": 64,
"n_heads": 2
}
},
"predictor": {
"sentence_dim": 64,
"n_style": 8,
"style_dim": 16,
"hdim": 128,
"n_layer": 2
}
}
}

223
Cs_HMI/SubProject/SuperTonic/assets/onnx/tts.yml Normal file
View File

@@ -0,0 +1,223 @@
tts_version: "v1.5.0"
split: "opensource-en"
ttl_ckpt_path: "unknown.pt"
dp_ckpt_path: "unknown.pt"
ae_ckpt_path: "unknown.pt"
ttl_train: "unknown"
dp_train: "unknown"
ae_train: "unknown"
ttl:
latent_dim: 24
chunk_compress_factor: 6
batch_expander:
n_batch_expand: 6
normalizer:
scale: 0.25
text_encoder:
char_dict_path: "resources/metadata/char_dict/opensource-en/char_dict.json"
text_embedder:
char_dict_path: "resources/metadata/char_dict/opensource-en/char_dict.json"
char_emb_dim: 256
convnext:
idim: 256
ksz: 5
intermediate_dim: 1024
num_layers: 6
dilation_lst: [1, 1, 1, 1, 1, 1]
attn_encoder:
hidden_channels: 256
filter_channels: 1024
n_heads: 4
n_layers: 4
p_dropout: 0.0
proj_out:
idim: 256
odim: 256
flow_matching:
sig_min: 0
style_encoder:
proj_in:
ldim: 24
chunk_compress_factor: 6
odim: 256
convnext:
idim: 256
ksz: 5
intermediate_dim: 1024
num_layers: 6
dilation_lst: [1, 1, 1, 1, 1, 1]
style_token_layer:
input_dim: 256
n_style: 50
style_key_dim: 256
style_value_dim: 256
prototype_dim: 256
n_units: 256
n_heads: 2
speech_prompted_text_encoder:
text_dim: 256
style_dim: 256
n_units: 256
n_heads: 2
uncond_masker:
prob_both_uncond: 0.04
prob_text_uncond: 0.01
std: 0.1
text_dim: 256
n_style: 50
style_key_dim: 256
style_value_dim: 256
vector_field:
proj_in:
ldim: 24
chunk_compress_factor: 6
odim: 512
time_encoder:
time_dim: 64
hdim: 256
main_blocks:
n_blocks: 4
time_cond_layer:
idim: 512
time_dim: 64
style_cond_layer:
idim: 512
style_dim: 256
text_cond_layer:
idim: 512
text_dim: 256
n_heads: 4
use_residual: True
rotary_base: 10000
rotary_scale: 10
convnext_0:
idim: 512
ksz: 5
intermediate_dim: 1024
num_layers: 4
dilation_lst: [1, 2, 4, 8]
convnext_1:
idim: 512
ksz: 5
intermediate_dim: 1024
num_layers: 1
dilation_lst: [1]
convnext_2:
idim: 512
ksz: 5
intermediate_dim: 1024
num_layers: 1
dilation_lst: [1]
last_convnext:
idim: 512
ksz: 5
intermediate_dim: 1024
num_layers: 4
dilation_lst: [1, 1, 1, 1]
proj_out:
idim: 512
chunk_compress_factor: 6
ldim: 24
ae:
sample_rate: 44100
n_delay: 0
base_chunk_size: 512
chunk_compress_factor: 1
ldim: 24
encoder:
spec_processor:
n_fft: 2048
win_length: 2048
hop_length: 512
n_mels: 228
sample_rate: 44100
eps: 1e-05
norm_mean: 0.0
norm_std: 1.0
ksz_init: 7
ksz: 7
num_layers: 10
dilation_lst: [1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
intermediate_dim: 2048
idim: 1253
hdim: 512
odim: 24
decoder:
ksz_init: 7
ksz: 7
num_layers: 10
dilation_lst: [1, 2, 4, 1, 2, 4, 1, 1, 1, 1]
intermediate_dim: 2048
idim: 24
hdim: 512
head:
idim: 512
hdim: 2048
odim: 512
ksz: 3
dp:
latent_dim: 24
chunk_compress_factor: 6
normalizer:
scale: 1.0
sentence_encoder:
char_emb_dim: 64
char_dict_path: "resources/metadata/char_dict/opensource-en/char_dict.json"
text_embedder:
char_dict_path: "resources/metadata/char_dict/opensource-en/char_dict.json"
char_emb_dim: 64
convnext:
idim: 64
ksz: 5
intermediate_dim: 256
num_layers: 6
dilation_lst: [1, 1, 1, 1, 1, 1]
attn_encoder:
hidden_channels: 64
filter_channels: 256
n_heads: 2
n_layers: 2
p_dropout: 0.0
proj_out:
idim: 64
odim: 64
style_encoder:
proj_in:
ldim: 24
chunk_compress_factor: 6
odim: 64
convnext:
idim: 64
ksz: 5
intermediate_dim: 256
num_layers: 4
dilation_lst: [1, 1, 1, 1]
style_token_layer:
input_dim: 64
n_style: 8
style_key_dim: 0
style_value_dim: 16
prototype_dim: 64
n_units: 64
n_heads: 2
predictor:
sentence_dim: 64
n_style: 8
style_dim: 16
hdim: 128
n_layer: 2
unicode_indexer_path: "/data/public/model/supertonic/tts/v1.5.0/opensource-en/onnx/unicode_indexer.npy"
unicode_indexer_json_path: "/data/public/model/supertonic/tts/v1.5.0/opensource-en/onnx/unicode_indexer.json"
window_path: "/data/public/model/supertonic/tts/v1.5.0/opensource-en/onnx/window.json"
filter_bank_path: "/data/public/model/supertonic/tts/v1.5.0/opensource-en/onnx/filter_bank.json"

1
Cs_HMI/SubProject/SuperTonic/assets/onnx/unicode_indexer.json Normal file
View File

File diff suppressed because one or more lines are too long

BIN
Cs_HMI/SubProject/SuperTonic/assets/onnx/vector_estimator.onnx Normal file
View File

Binary file not shown.

BIN
Cs_HMI/SubProject/SuperTonic/assets/onnx/vocoder.onnx Normal file
View File

Binary file not shown.

13076
Cs_HMI/SubProject/SuperTonic/assets/voice_styles/F1.json Normal file
View File

File diff suppressed because it is too large Load Diff

13076
Cs_HMI/SubProject/SuperTonic/assets/voice_styles/F2.json Normal file
View File

File diff suppressed because it is too large Load Diff

13076
Cs_HMI/SubProject/SuperTonic/assets/voice_styles/M1.json Normal file
View File

File diff suppressed because it is too large Load Diff

13076
Cs_HMI/SubProject/SuperTonic/assets/voice_styles/M2.json Normal file
View File

File diff suppressed because it is too large Load Diff

15
Cs_HMI/SubProject/SuperTonic/packages.config Normal file
View File

@@ -0,0 +1,15 @@
<?xml version="1.0" encoding="utf-8"?>
<packages>
<package id="Microsoft.Bcl.AsyncInterfaces" version="10.0.1" targetFramework="net48" />
<package id="Microsoft.ML.OnnxRuntime" version="1.23.2" targetFramework="net48" />
<package id="Microsoft.ML.OnnxRuntime.Managed" version="1.23.2" targetFramework="net48" />
<package id="System.Buffers" version="4.6.1" targetFramework="net48" />
<package id="System.IO.Pipelines" version="10.0.1" targetFramework="net48" />
<package id="System.Memory" version="4.6.3" targetFramework="net48" />
<package id="System.Numerics.Vectors" version="4.6.1" targetFramework="net48" />
<package id="System.Runtime.CompilerServices.Unsafe" version="6.1.2" targetFramework="net48" />
<package id="System.Text.Encodings.Web" version="10.0.1" targetFramework="net48" />
<package id="System.Text.Json" version="10.0.1" targetFramework="net48" />
<package id="System.Threading.Tasks.Extensions" version="4.6.3" targetFramework="net48" />
<package id="System.ValueTuple" version="4.6.1" targetFramework="net48" />
</packages>