// /// <summary>
// /// IMPORTANT: This will crash the Lipsyncer.
// ///
// /// Get this. I only have EnglishUS and EnglishGB recognizers for my OS
// /// (win7 pro) but can output the phonemes of French and perhaps other
// /// languages.
// /// PS. My EnglishGB phonemes appear to be utterly borked.
// /// </summary>
// /// <param name="langid"></param>
// void PrintPhons(int langid)
// {
// _phoneConverter.LanguageId = langid;
// for (int i = 0; i != 100; ++i)
// logfile.Log(i + " - " + _phoneConverter.IdToPhone(i));
// }
#region methods
/// <summary>
/// Sets the Recognizer and LanguageId when the Recognizers combobox
/// selection changes.
/// @note The LanguageId is used by both TTS and SpeechRecognition.
/// </summary>
/// <param name="recognizer"></param>
/// <returns>true if the language-id is set successfully</returns>
bool SetRecognizer(Recognizer recognizer)
{
#if DEBUG
logfile.Log();
logfile.Log("SetRecognizer()");
logfile.Log(". create (SpInprocRecognizer)_recognizer");
#endif
_recognizer = new SpInprocRecognizer();
#if DEBUG
logfile.Log(". (SpInprocRecognizer)_recognizer CREATED");
#endif
_recognizer.Recognizer = (SpObjectToken)recognizer.Tok;
#if DEBUG
logfile.Log(". recognizer.Tok.Id= " + recognizer.Tok.Id);
logfile.Log(". recognizer.Description= " + recognizer.Tok.GetDescription());
logfile.Log(". recognizer.Langids= " + recognizer.Langids);
#endif
string langid = recognizer.Langids;
int pos = recognizer.Langids.IndexOf(' ');
if (pos != -1)
{
langid = langid.Substring(0, pos); // use 1st langid
}
// TODO: ComboBox dropdown for user to choose from if 2+ languages
// are supported by the current Recognizer.
int id;
if (!Int32.TryParse(langid, out id) || // safety - unless the token has "n/a" Languages.
id < 0) // TODO: check id against valid SAPI language-ids
{
if (!FxeGeneratorF.isConsole)
{
using (var d = new InfoDialog("Error", "Did not find a Language in the Recognizer's token."))
{
d.ShowDialog(FxeGeneratorF.That);
}
}
return(false);
}
_phoneConverter.LanguageId = id;
#if DEBUG
logfile.Log(". _phoneConverter.LanguageId= " + _phoneConverter.LanguageId);
logfile.Log();
#endif
StaticData.viceroy(_phoneConverter.LanguageId);
return(true);
}
/// <summary>
/// Tests an FXE-file by reading it w/ <see cref="FxeReader.TestFile"/>.
/// </summary>
/// <param name="pfe">fullpath of FXE-file to test</param>
static void TestOutputFile(string pfe)
{
string titl, info;
if (FxeReader.TestFile(pfe))
{
titl = "Write SUCCESS";
info = pfe;
}
else
{
titl = "Write FAILED";
info = "Borked file" + Environment.NewLine + pfe;
}
using (var d = new InfoDialog(titl, info))
{
d.ShowDialog(FxeGeneratorF.That);
}
}
/// <summary>
/// cTor.
/// </summary>
/// <param name="pfe">blank string if '!isConsole'</param>
/// <param name="headtype">blank string if '!isConsole'</param>
internal FxeGeneratorF(string pfe = "", string headtype = "")
{
#if DEBUG
// LogSpeechRecognitionEngines();
logfile.Log("FxeGeneratorF() cTor pfe= " + pfe + " headtype= " + headtype);
#endif
That = this;
FxeData.LoadTrigrams();
bool fatality = false;
if (pfe == String.Empty) // is GUI interface ->
{
#if DEBUG
logfile.Log(". is GUI");
#endif
InitializeComponent();
co_headtype.SelectedIndex = 0;
la_def_word_pct.Text =
la_def_phon_pct.Text =
la_enh_word_pct.Text =
la_enh_phon_pct.Text = String.Empty;
tb_text.Text = String.Empty;
tb_text.BackColor =
tb_expected.BackColor = Color.AntiqueWhite;
tb_def_words.BackColor =
tb_def_phons.BackColor =
tb_enh_words.BackColor =
tb_enh_phons.BackColor = Color.GhostWhite;
// PHONEMES data/grid ->
DataColumn dc;
dc = new DataColumn(HEAD_PHONS_0, typeof(string)); // pos
dc.ReadOnly = true;
_dt1.Columns.Add(dc);
dc = new DataColumn(HEAD_PHONS_1, typeof(string)); // phon
dc.ReadOnly = true;
_dt1.Columns.Add(dc);
dc = new DataColumn(HEAD_PHONS_2, typeof(string)); // start
dc.ReadOnly = true;
_dt1.Columns.Add(dc);
dc = new DataColumn(HEAD_PHONS_3, typeof(string)); // stop
dc.ReadOnly = true;
_dt1.Columns.Add(dc);
dc = new DataColumn(HEAD_PHONS_4, typeof(string)); // vis
dc.ReadOnly = true;
_dt1.Columns.Add(dc);
dc = new DataColumn(HEAD_PHONS_5, typeof(string)); // truth
dc.ReadOnly = true;
_dt1.Columns.Add(dc);
dc = new DataColumn(HEAD_PHONS_6, typeof(string)); // level
dc.ReadOnly = true;
_dt1.Columns.Add(dc);
grid_phons.DataSource = _dt1;
grid_phons.Columns[0].Width = 50; // 50 "pos"
grid_phons.Columns[1].Width = 76; // 76 "phoneme"
grid_phons.Columns[2].Width = 86; // 86 "start"
grid_phons.Columns[3].Width = 86; // 86 "stop"
grid_phons.Columns[4].Width = 67; // 67 "viseme"
grid_phons.Columns[5].Width = 57; // 57 "truth"
grid_phons.Columns[6].Width = 61; // 56 "level"
for (int i = 0; i != grid_phons.Columns.Count; ++i)
{
grid_phons.Columns[i].SortMode = DataGridViewColumnSortMode.NotSortable;
}
// Data Blocks data/grid ->
dc = new DataColumn(HEAD_BLOCKS_0, typeof(string));
dc.ReadOnly = true;
_dt2.Columns.Add(dc);
dc = new DataColumn(HEAD_BLOCKS_1, typeof(float));
dc.ReadOnly = true;
_dt2.Columns.Add(dc);
dc = new DataColumn(HEAD_BLOCKS_2, typeof(float));
dc.ReadOnly = true;
_dt2.Columns.Add(dc);
grid_blocs.DataSource = _dt2;
// grid_blocs.Columns[0].Width = 80; // 68 "viseme"
// grid_blocs.Columns[1].Width = 97; // 87 "frame stop"
// grid_blocs.Columns[2].Width = 110; // 100 "morph weight"
grid_blocs.Columns[0].Width = 140; // 68 "viseme"
grid_blocs.Columns[1].Width = 67; // "stop"
grid_blocs.Columns[2].Width = 80; // "weight"
// for (int i = 0; i != grid_blocs.Columns.Count; ++i)
// grid_blocs.Columns[i].SortMode = DataGridViewColumnSortMode.NotSortable;
grid_blocs.Columns[0].SortMode = DataGridViewColumnSortMode.NotSortable;
grid_blocs.Columns[2].SortMode = DataGridViewColumnSortMode.NotSortable;
grid_blocs.ColumnHeaderMouseClick += dgblocs_ColumnHeaderMouseClick;
grid_phons.RowHeadersVisible =
grid_blocs.RowHeadersVisible = false;
printversion();
// instantiate/initialize SAPI
_sapi = new SapiLipsync();
_sapi.TtsStreamEnded += OnTtsStreamEnded;
_sapi.SrStreamEnded += OnSrStreamEnded;
// this will set '_sapi._recognizer'
// this will set '_sapi._phoneConverter.LanguageId'
// and the Titletext
if (!SpeechRecognizerLister.AddSpeechRecognizers(co_recognizers))
{
string info = "FXE LipSyncer requires a SAPI 5.4 compliant" + Environment.NewLine
+ "Microsoft Speech Recognizer" + Environment.NewLine
+ "as displayed in Windoz ControlPanel|SpeechRecognition." + Environment.NewLine
+ Environment.NewLine
+ "none was found ...";
using (var d = new InfoDialog("FATAL Error", info))
{
d.ShowDialog(this);
}
fatality = true;
}
}
else if (headtype != String.Empty && File.Exists(pfe)) // is Console interface ->
{
#if DEBUG
logfile.Log(". is Console");
#endif
// TODO: Fail if a Recognizer is not found.
// TODO: Ensure that 'head Model/Skeleton type' is a recognized type.
// Eg. "P_HHM"
Filelabel = Utility.GetFilelabel(pfe); // NOTE: that will be written into the FXE-file output.
_sapi = new SapiLipsync(_pfe = pfe);
if (_sapi.Wavefile != String.Empty)
{
isConsole = true;
_headtype = headtype;
_sapi.SrStreamEnded += OnSrStreamEnded;
_sapi.Start(LoadTypedTextFile());
}
else
{
fatality = true;
}
}
else // is Console error ->
{
fatality = true;
}
if (fatality)
{
Environment.Exit(0);
}
}
/// <summary>
/// Determines the file to use for the SpeechRecognition filestream
/// converting it from BMU/MP3 to WAV if necessary.
/// @note The result shall be PCM 44.1kHz 16-bit Mono.
/// </summary>
/// <param name="pfe">path_file_extension</param>
/// <returns>the fullpath to a PCM-wave file else a blank-string</returns>
internal static string deterwave(string pfe)
{
//logfile.Log("AudioConverter.deterAudiopath() pfe= " + pfe);
string pathT = Path.GetTempPath();
//logfile.Log(". path= " + pathT);
if (pfe.EndsWith(EXT_WAV, StringComparison.InvariantCultureIgnoreCase) || // prep .BMU ->
pfe.EndsWith(EXT_BMU, StringComparison.InvariantCultureIgnoreCase))
{
var chars = new char[3];
using (var fs = new FileStream(pfe, FileMode.Open, FileAccess.Read, FileShare.Read))
{
var br = new BinaryReader(fs);
chars = br.ReadChars(3);
br.Close();
}
if (chars[0] == 'B' && // because .BMUs are .MP3s and NwN2 labels them as .WAVs
chars[1] == 'M' &&
chars[2] == 'U')
{
string pfeT = Path.Combine(pathT, TMP_MP3); // so label it as .MP3 and allow the next block to catch it.
//logfile.Log(". pfeT(MP3)= " + pfeT);
File.Delete(pfeT);
File.Copy(pfe, pfeT);
pfe = pfeT;
}
}
if (pfe.EndsWith(EXT_MP3, StringComparison.InvariantCultureIgnoreCase)) // convert to .WAV file ->
{
string pfeT = Path.Combine(pathT, TMP_WAV);
//logfile.Log(". pfeT(WAV)= " + pfeT);
File.Delete(pfeT);
// string execpath = Path.GetDirectoryName(Assembly.GetExecutingAssembly().Location);
// var info = new ProcessStartInfo(Path.Combine(execpath, LAME_EXE));
var info = new ProcessStartInfo(Path.Combine(Application.StartupPath, LAME_EXE));
info.Arguments = "--decode \"" + pfe + "\" \"" + pfeT + "\"";
info.WindowStyle = ProcessWindowStyle.Hidden;
info.UseShellExecute = false;
info.CreateNoWindow = true;
using (Process proc = Process.Start(info))
{
proc.WaitForExit();
}
pfe = pfeT;
}
// http://www.topherlee.com/software/pcm-tut-wavformat.html
// 1- 4 "RIFF" Marks the file as a riff file. Characters are each 1 byte long.
// 5- 8 File size (integer) Size of the overall file - 8 bytes, in bytes (32-bit integer). Typically, you'd fill this in after creation.
// 9-12 "WAVE" File Type Header. For our purposes, it always equals "WAVE".
// 13-16 "fmt " Format chunk marker. Includes trailing null
// 17-20 16 Length of format data as listed above
// 21-22 1 Type of format (1 is PCM) - 2 byte integer
// 23-24 2 Number of Channels - 2 byte integer
// 25-28 44100 Sample Rate - 32 byte integer. Common values are 44100 (CD), 48000 (DAT). Sample Rate = Number of Samples per second, or Hertz.
// 29-32 176400 (Sample Rate * BitsPerSample * Channels) / 8.
// 33-34 4 (BitsPerSample * Channels) / 8.1 - 8 bit mono2 - 8 bit stereo/16 bit mono4 - 16 bit stereo
// 35-36 16 Bits per sample
// 37-40 "data" "data" chunk header. Marks the beginning of the data section.
// 41-44 File size (data) Size of the data section.
string fullpath = String.Empty;
if (pfe.EndsWith(EXT_WAV, StringComparison.InvariantCultureIgnoreCase)) // check .WAV ->
{
using (var fs = new FileStream(pfe, FileMode.Open, FileAccess.Read, FileShare.Read))
{
var br = new BinaryReader(fs);
char[] c = br.ReadChars(16); // start 0
if (c[0] == 'R' && c[1] == 'I' && c[2] == 'F' && c[3] == 'F' &&
c[8] == 'W' && c[9] == 'A' && c[10] == 'V' && c[11] == 'E' &&
c[12] == 'f' && c[13] == 'm' && c[14] == 't' && c[15] == ' ')
{
br.ReadBytes(4); // start 16
short format = br.ReadInt16(); // start 20: is PCM
if (format == (short)1)
{
short channels = br.ReadInt16(); // start 22: is Mono
if (channels == (short)1)
{
// TODO: Sample-rate and bit-depth should probably be relaxed.
int rate = br.ReadInt32(); // start 24: is 44.1kHz
if (rate == 44100)
{
br.ReadBytes(6); // start 28
short bits = br.ReadInt16(); // start 34: is 16-bit
if (bits == (short)16)
{
fullpath = pfe;
//logfile.Log(". AudioConverter.fullpath= " + fullpath);
}
}
}
}
}
br.Close();
}
}
if (!FxeGeneratorF.isConsole && fullpath == String.Empty)
{
using (var d = new InfoDialog("Conversion Error", "Failed to convert to 44.1kHz 16-bit Mono PCM-wave format."))
{
d.ShowDialog(FxeGeneratorF.That);
}
}
return(fullpath);
}