experiments/STT
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Match native error codes for .NET

Browse Source
This commit is contained in:
Carlos Fonseca Murillo 2019-05-29 08:08:35 +00:00
parent df5545299e
commit 73511e48b2
10 changed files with 151 additions and 94 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
examples/net_framework/DeepSpeechWPF/DeepSpeech.WPF.csproj
View File

@ -14,26 +14,8 @@
<WarningLevel>4</WarningLevel>
<AutoGenerateBindingRedirects>true</AutoGenerateBindingRedirects>
<Deterministic>true</Deterministic>
</PropertyGroup>
<PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Debug|AnyCPU' ">
<PlatformTarget>AnyCPU</PlatformTarget>
<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' ">
<PlatformTarget>AnyCPU</PlatformTarget>
<DebugType>pdbonly</DebugType>
<Optimize>true</Optimize>
<OutputPath>bin\Release\</OutputPath>
<DefineConstants>TRACE</DefineConstants>
<ErrorReport>prompt</ErrorReport>
<WarningLevel>4</WarningLevel>
<AllowUnsafeBlocks>true</AllowUnsafeBlocks>
<NuGetPackageImportStamp>
</NuGetPackageImportStamp>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)|$(Platform)' == 'Debug|x64'">
<DebugSymbols>true</DebugSymbols>
@ -59,13 +41,14 @@
</PropertyGroup>
<ItemGroup>
<Reference Include="CSCore, Version=1.2.1.2, Culture=neutral, PublicKeyToken=5a08f2b6f4415dea, processorArchitecture=MSIL">
<HintPath>..\packages\CSCore.1.2.1.2\lib\net35-client\CSCore.dll</HintPath>
<HintPath>packages\CSCore.1.2.1.2\lib\net35-client\CSCore.dll</HintPath>
</Reference>
<Reference Include="NAudio, Version=1.8.5.0, Culture=neutral, processorArchitecture=MSIL">
<HintPath>..\packages\NAudio.1.8.5\lib\net35\NAudio.dll</HintPath>
<Reference Include="NAudio, Version=1.9.0.0, Culture=neutral, processorArchitecture=MSIL">
<HintPath>packages\NAudio.1.9.0\lib\net35\NAudio.dll</HintPath>
</Reference>
<Reference Include="System" />
<Reference Include="System.Data" />
<Reference Include="System.Windows.Forms" />
<Reference Include="System.Xml" />
<Reference Include="Microsoft.CSharp" />
<Reference Include="System.Core" />
@ -125,7 +108,7 @@
<None Include="App.config" />
</ItemGroup>
<ItemGroup>
<ProjectReference Include="..\DeepSpeechClient\DeepSpeechClient.csproj">
<ProjectReference Include="..\..\..\native_client\dotnet\DeepSpeechClient\DeepSpeechClient.csproj">
<Project>{56de4091-bbbe-47e4-852d-7268b33b971f}</Project>
<Name>DeepSpeechClient</Name>
</ProjectReference>

31
examples/net_framework/DeepSpeechWPF/DeepSpeech.WPF.sln Normal file
View File

@ -0,0 +1,31 @@

Microsoft Visual Studio Solution File, Format Version 12.00
# Visual Studio 15
VisualStudioVersion = 15.0.28307.421
MinimumVisualStudioVersion = 10.0.40219.1
Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "DeepSpeech.WPF", "DeepSpeech.WPF.csproj", "{54BFD766-4305-4F4C-BA59-AF45505DF3C1}"
EndProject
Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "DeepSpeechClient", "..\..\..\native_client\dotnet\DeepSpeechClient\DeepSpeechClient.csproj", "{56DE4091-BBBE-47E4-852D-7268B33B971F}"
EndProject
Global
GlobalSection(SolutionConfigurationPlatforms) = preSolution
Debug|x64 = Debug|x64
Release|x64 = Release|x64
EndGlobalSection
GlobalSection(ProjectConfigurationPlatforms) = postSolution
{54BFD766-4305-4F4C-BA59-AF45505DF3C1}.Debug|x64.ActiveCfg = Debug|x64
{54BFD766-4305-4F4C-BA59-AF45505DF3C1}.Debug|x64.Build.0 = Debug|x64
{54BFD766-4305-4F4C-BA59-AF45505DF3C1}.Release|x64.ActiveCfg = Release|x64
{54BFD766-4305-4F4C-BA59-AF45505DF3C1}.Release|x64.Build.0 = Release|x64
{56DE4091-BBBE-47E4-852D-7268B33B971F}.Debug|x64.ActiveCfg = Debug|x64
{56DE4091-BBBE-47E4-852D-7268B33B971F}.Debug|x64.Build.0 = Debug|x64
{56DE4091-BBBE-47E4-852D-7268B33B971F}.Release|x64.ActiveCfg = Release|x64
{56DE4091-BBBE-47E4-852D-7268B33B971F}.Release|x64.Build.0 = Release|x64
EndGlobalSection
GlobalSection(SolutionProperties) = preSolution
HideSolutionNode = FALSE
EndGlobalSection
GlobalSection(ExtensibilityGlobals) = postSolution
SolutionGuid = {19C58802-CCEC-4FD1-8D17-A6EB766116F7}
EndGlobalSection
EndGlobal

22
examples/net_framework/DeepSpeechWPF/MainWindow.xaml.cs
View File

@ -79,14 +79,8 @@ namespace DeepSpeechWPF
{
try
{
if (_sttClient.CreateModel("output_graph.pbmm", N_CEP, N_CONTEXT, "alphabet.txt", BEAM_WIDTH) == 0)
{
Dispatcher.Invoke(() => { EnableControls(); });
}
else
{
MessageBox.Show("Model load failed.");
}
_sttClient.CreateModel("output_graph.pbmm", N_CEP, N_CONTEXT, "alphabet.txt", BEAM_WIDTH);
Dispatcher.Invoke(() => { EnableControls(); });
}
catch (Exception ex)
{
@ -161,18 +155,12 @@ namespace DeepSpeechWPF
{
try
{
if (_sttClient.EnableDecoderWithLM("alphabet.txt", "lm.binary", "trie", LM_ALPHA, LM_BETA) != 0)
{
MessageBox.Show("Error loading LM.");
Dispatcher.Invoke(() => btnEnableLM.IsEnabled = true);
}
else
{
Dispatcher.Invoke(() => lblStatus.Content = "LM loaded.");
}
_sttClient.EnableDecoderWithLM("alphabet.txt", "lm.binary", "trie", LM_ALPHA, LM_BETA);
Dispatcher.Invoke(() => lblStatus.Content = "LM loaded.");
}
catch (Exception ex)
{
Dispatcher.Invoke(() => btnEnableLM.IsEnabled = true);
MessageBox.Show(ex.Message);
}
});

2
examples/net_framework/DeepSpeechWPF/packages.config
View File

@ -1,5 +1,5 @@
<?xml version="1.0" encoding="utf-8"?>
<packages>
<package id="CSCore" version="1.2.1.2" targetFramework="net462" />
<package id="NAudio" version="1.8.5" targetFramework="net462" />
<package id="NAudio" version="1.9.0" targetFramework="net462" />
</packages>

62
native_client/dotnet/DeepSpeechClient/DeepSpeech.cs
View File

@ -5,6 +5,7 @@ using DeepSpeechClient.Extensions;
using System;
using System.IO;
using System.Runtime.InteropServices;
using DeepSpeechClient.Enums;
namespace DeepSpeechClient
{
@ -35,8 +36,8 @@ namespace DeepSpeechClient
/// <param name="aNContext">The context window the model was trained with.</param>
/// <param name="aAlphabetConfigPath">The path to the configuration file specifying the alphabet used by the network.</param>
/// <param name="aBeamWidth">The beam width used by the decoder. A larger beam width generates better results at the cost of decoding time.</param>
/// <returns>Zero on success, non-zero on failure.</returns>
public unsafe int CreateModel(string aModelPath, uint aNCep,
/// <exception cref="ArgumentException">Thrown when the native binary failed to create the model.</exception>
public unsafe void CreateModel(string aModelPath, uint aNCep,
uint aNContext, string aAlphabetConfigPath, uint aBeamWidth)
{
string exceptionMessage = null;
@ -61,16 +62,53 @@ namespace DeepSpeechClient
{
throw new FileNotFoundException(exceptionMessage);
}
int result = NativeImp.DS_CreateModel(aModelPath,
var resultCode = NativeImp.DS_CreateModel(aModelPath,
aNCep,
aNContext,
aAlphabetConfigPath,
aBeamWidth,
ref _modelStatePP);
EvaluateResultCode(resultCode);
_modelStateP = *_modelStatePP;
return result;
}
/// <summary>
/// Evaluate the result code and will raise an exception if necessary.
/// </summary>
/// <param name="resultCode">Native result code.</param>
private void EvaluateResultCode(ErrorCodes resultCode)
{
switch (resultCode)
{
case ErrorCodes.DS_ERR_OK:
break;
case ErrorCodes.DS_ERR_NO_MODEL:
throw new ArgumentException("Missing model information.");
case ErrorCodes.DS_ERR_INVALID_ALPHABET:
throw new ArgumentException("Invalid alphabet file or invalid alphabet size.");
case ErrorCodes.DS_ERR_INVALID_SHAPE:
throw new ArgumentException("Invalid model shape.");
case ErrorCodes.DS_ERR_INVALID_LM:
throw new ArgumentException("Invalid language model file.");
case ErrorCodes.DS_ERR_FAIL_INIT_MMAP:
throw new ArgumentException("Failed to initialize memory mapped model.");
case ErrorCodes.DS_ERR_FAIL_INIT_SESS:
throw new ArgumentException("Failed to initialize the session.");
case ErrorCodes.DS_ERR_FAIL_INTERPRETER:
throw new ArgumentException("Interpreter failed.");
case ErrorCodes.DS_ERR_FAIL_RUN_SESS:
throw new ArgumentException("Failed to run the session.");
case ErrorCodes.DS_ERR_FAIL_CREATE_STREAM:
throw new ArgumentException("Error creating the stream.");
case ErrorCodes.DS_ERR_FAIL_READ_PROTOBUF:
throw new ArgumentException("Error reading the proto buffer model file.");
case ErrorCodes.DS_ERR_FAIL_CREATE_SESS:
throw new ArgumentException("Error failed to create session.");
case ErrorCodes.DS_ERR_MODEL_INCOMPATIBLE:
throw new ArgumentException("Error incompatible model.");
default:
throw new ArgumentException("Unknown error, please make sure you are using the correct native binary.");
}
}
/// <summary>
@ -89,8 +127,8 @@ namespace DeepSpeechClient
/// <param name="aTriePath">The path to the trie file build from the same vocabulary as the language model binary.</param>
/// <param name="aLMAlpha">The alpha hyperparameter of the CTC decoder. Language Model weight.</param>
/// <param name="aLMBeta">The beta hyperparameter of the CTC decoder. Word insertion weight.</param>
/// <returns>Zero on success, non-zero on failure (invalid arguments).</returns>
public unsafe int EnableDecoderWithLM(string aAlphabetConfigPath,
/// <exception cref="ArgumentException">Thrown when the native binary failed to enable decoding with a language model.</exception>
public unsafe void EnableDecoderWithLM(string aAlphabetConfigPath,
string aLMPath, string aTriePath,
float aLMAlpha, float aLMBeta)
{
@ -109,12 +147,13 @@ namespace DeepSpeechClient
throw new FileNotFoundException(exceptionMessage);
}
return NativeImp.DS_EnableDecoderWithLM(_modelStatePP,
var resultCode = NativeImp.DS_EnableDecoderWithLM(_modelStatePP,
aAlphabetConfigPath,
aLMPath,
aTriePath,
aLMAlpha,
aLMBeta);
EvaluateResultCode(resultCode);
}
/// <summary>
@ -169,10 +208,11 @@ namespace DeepSpeechClient
/// One timestep is equivalent to two window lengths(20ms).
/// If set to 0 we reserve enough frames for 3 seconds of audio(150).</param>
/// <param name="aSampleRate">The sample-rate of the audio signal</param>
/// <returns>Zero for success, non-zero on failure</returns>
public unsafe int SetupStream(uint aPreAllocFrames, uint aSampleRate)
/// <exception cref="ArgumentException">Thrown when the native binary failed to initialize the streaming mode.</exception>
public unsafe void SetupStream(uint aPreAllocFrames, uint aSampleRate)
{
return NativeImp.DS_SetupStream(_modelStatePP, aPreAllocFrames, aSampleRate, ref _streamingStatePP);
var resultCode = NativeImp.DS_SetupStream(_modelStatePP, aPreAllocFrames, aSampleRate, ref _streamingStatePP);
EvaluateResultCode(resultCode);
}
/// <summary>

1
native_client/dotnet/DeepSpeechClient/DeepSpeechClient.csproj
View File

@ -45,6 +45,7 @@
</ItemGroup>
<ItemGroup>
<Compile Include="DeepSpeech.cs" />
<Compile Include="Enums\ErrorCodes.cs" />
<Compile Include="Interfaces\IDeepSpeech.cs" />
<Compile Include="Extensions\NativeExtensions.cs" />
<Compile Include="Models\Metadata.cs" />

29
native_client/dotnet/DeepSpeechClient/Enums/ErrorCodes.cs Normal file
View File

@ -0,0 +1,29 @@
namespace DeepSpeechClient.Enums
{
/// <summary>
/// Error codes from the native DeepSpeech binary.
/// </summary>
internal enum ErrorCodes
{
// OK
DS_ERR_OK = 0x0000,
// Missing invormations
DS_ERR_NO_MODEL = 0x1000,
// Invalid parameters
DS_ERR_INVALID_ALPHABET = 0x2000,
DS_ERR_INVALID_SHAPE = 0x2001,
DS_ERR_INVALID_LM = 0x2002,
DS_ERR_MODEL_INCOMPATIBLE = 0x2003,
// Runtime failures
DS_ERR_FAIL_INIT_MMAP = 0x3000,
DS_ERR_FAIL_INIT_SESS = 0x3001,
DS_ERR_FAIL_INTERPRETER = 0x3002,
DS_ERR_FAIL_RUN_SESS = 0x3003,
DS_ERR_FAIL_CREATE_STREAM = 0x3004,
DS_ERR_FAIL_READ_PROTOBUF = 0x3005,
DS_ERR_FAIL_CREATE_SESS = 0x3006,
}
}

12
native_client/dotnet/DeepSpeechClient/Interfaces/IDeepSpeech.cs
View File

@ -21,8 +21,8 @@ namespace DeepSpeechClient.Interfaces
/// <param name="aNContext">The context window the model was trained with.</param>
/// <param name="aAlphabetConfigPath">The path to the configuration file specifying the alphabet used by the network.</param>
/// <param name="aBeamWidth">The beam width used by the decoder. A larger beam width generates better results at the cost of decoding time.</param>
/// <returns>Zero on success, non-zero on failure.</returns>
unsafe int CreateModel(string aModelPath, uint aNCep,
/// <exception cref="ArgumentException">Thrown when the native binary failed to create the model.</exception>
unsafe void CreateModel(string aModelPath, uint aNCep,
uint aNContext,
string aAlphabetConfigPath,
uint aBeamWidth);
@ -35,8 +35,8 @@ namespace DeepSpeechClient.Interfaces
/// <param name="aTriePath">The path to the trie file build from the same vocabulary as the language model binary.</param>
/// <param name="aLMAlpha">The alpha hyperparameter of the CTC decoder. Language Model weight.</param>
/// <param name="aLMBeta">The beta hyperparameter of the CTC decoder. Word insertion weight.</param>
/// <returns>Zero on success, non-zero on failure (invalid arguments).</returns>
unsafe int EnableDecoderWithLM(string aAlphabetConfigPath,
/// <exception cref="ArgumentException">Thrown when the native binary failed to enable decoding with a language model.</exception>
unsafe void EnableDecoderWithLM(string aAlphabetConfigPath,
string aLMPath,
string aTriePath,
float aLMAlpha,
@ -88,8 +88,8 @@ namespace DeepSpeechClient.Interfaces
/// One timestep is equivalent to two window lengths(20ms).
/// If set to 0 we reserve enough frames for 3 seconds of audio(150).</param>
/// <param name="aSampleRate">The sample-rate of the audio signal</param>
/// <returns>Zero for success, non-zero on failure</returns>
unsafe int SetupStream(uint aPreAllocFrames, uint aSampleRate);
/// <exception cref="ArgumentException">Thrown when the native binary failed to initialize the streaming mode.</exception>
unsafe void SetupStream(uint aPreAllocFrames, uint aSampleRate);
/// <summary>
/// Feeds audio samples to an ongoing streaming inference.

9
native_client/dotnet/DeepSpeechClient/NativeImp.cs
View File

@ -1,4 +1,5 @@
using DeepSpeechClient.Structs;
using DeepSpeechClient.Enums;
using DeepSpeechClient.Structs;
using System;
using System.Runtime.InteropServices;
@ -15,7 +16,7 @@ namespace DeepSpeechClient
internal static extern void DS_PrintVersions();
[DllImport("libdeepspeech.so", CallingConvention = CallingConvention.Cdecl)]
internal unsafe static extern int DS_CreateModel(string aModelPath,
internal unsafe static extern ErrorCodes DS_CreateModel(string aModelPath,
uint aNCep,
uint aNContext,
string aAlphabetConfigPath,
@ -23,7 +24,7 @@ namespace DeepSpeechClient
ref ModelState** pint);
[DllImport("libdeepspeech.so", CallingConvention = CallingConvention.Cdecl)]
internal static unsafe extern int DS_EnableDecoderWithLM(ModelState** aCtx,
internal static unsafe extern ErrorCodes DS_EnableDecoderWithLM(ModelState** aCtx,
string aAlphabetConfigPath,
string aLMPath,
string aTriePath,
@ -47,7 +48,7 @@ namespace DeepSpeechClient
internal static unsafe extern void DS_DestroyModel(ModelState** aCtx);
[DllImport("libdeepspeech.so", CallingConvention = CallingConvention.Cdecl)]
internal static unsafe extern int DS_SetupStream(ModelState** aCtx,
internal static unsafe extern ErrorCodes DS_SetupStream(ModelState** aCtx,
uint aPreAllocFrames,
uint aSampleRate, ref StreamingState** retval);

46
native_client/dotnet/DeepSpeechConsole/Program.cs
View File

@ -25,7 +25,7 @@ namespace CSharpExamples
{
var nl = Environment.NewLine;
string retval =
Environment.NewLine +$"Recognized text: {string.Join("", meta?.Items?.Select(x=>x.Character))} {nl}"
Environment.NewLine + $"Recognized text: {string.Join("", meta?.Items?.Select(x => x.Character))} {nl}"
+ $"Prob: {meta?.Probability} {nl}"
+ $"Item count: {meta?.Items?.Length} {nl}"
+ string.Join(nl, meta?.Items?.Select(x => $"Timestep : {x.Timestep} TimeOffset: {x.StartTime} Char: {x.Character}"));
@ -60,43 +60,27 @@ namespace CSharpExamples
using (IDeepSpeech sttClient = new DeepSpeech())
{
var result = 1;
Console.WriteLine("Loading model...");
stopwatch.Start();
try
{
result = sttClient.CreateModel(
Console.WriteLine("Loading model...");
stopwatch.Start();
sttClient.CreateModel(
model ?? "output_graph.pbmm",
N_CEP, N_CONTEXT,
alphabet ?? "alphabet.txt",
BEAM_WIDTH);
}
catch (IOException ex)
{
Console.WriteLine("Error loading lm.");
Console.WriteLine(ex.Message);
}
stopwatch.Stop();
if (result == 0)
{
stopwatch.Stop();
Console.WriteLine($"Model loaded - {stopwatch.Elapsed.Milliseconds} ms");
stopwatch.Reset();
if (lm != null)
{
Console.WriteLine("Loadin LM...");
try
{
result = sttClient.EnableDecoderWithLM(
alphabet ?? "alphabet.txt",
lm ?? "lm.binary",
trie ?? "trie",
LM_ALPHA, LM_BETA);
}
catch (IOException ex)
{
Console.WriteLine("Error loading lm.");
Console.WriteLine(ex.Message);
}
sttClient.EnableDecoderWithLM(
alphabet ?? "alphabet.txt",
lm ?? "lm.binary",
trie ?? "trie",
LM_ALPHA, LM_BETA);
}
@ -123,15 +107,15 @@ namespace CSharpExamples
Console.WriteLine($"Audio duration: {waveInfo.TotalTime.ToString()}");
Console.WriteLine($"Inference took: {stopwatch.Elapsed.ToString()}");
Console.WriteLine((extended ? $"Extended result: ": "Recognized text: ") + speechResult);
Console.WriteLine((extended ? $"Extended result: " : "Recognized text: ") + speechResult);
}
waveBuffer.Clear();
}
else
catch (Exception ex)
{
Console.WriteLine("Error loding the model.");
Console.WriteLine(ex.Message);
}
}
}
}
}
}