EngineX-Iluvatar/enginex-mr_series-sherpa-onnx
Archived
16
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

Add C# API for spoken language identification (#697)

Browse Source
This commit is contained in:
Fangjun Kuang
2024-03-25 18:45:09 +08:00
committed by GitHub
parent 83a10a55a5
commit 305c373107
10 changed files with 265 additions and 55 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
dotnet-examples/sherpa-onnx.sln
View File

@@ -13,6 +13,8 @@ Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "offline-tts", "offline-tts\
EndProject
Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "offline-tts-play", "offline-tts-play\offline-tts-play.csproj", "{40781464-5948-462B-BA4B-98932711513F}"
EndProject
Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "spoken-language-identification", "spoken-language-identification\spoken-language-identification.csproj", "{3D7CF3D6-AC45-4D50-9619-5687B1443E94}"
EndProject
Global
GlobalSection(SolutionConfigurationPlatforms) = preSolution
Debug|Any CPU = Debug|Any CPU
@@ -42,5 +44,9 @@ Global
{40781464-5948-462B-BA4B-98932711513F}.Debug|Any CPU.Build.0 = Debug|Any CPU
{40781464-5948-462B-BA4B-98932711513F}.Release|Any CPU.ActiveCfg = Release|Any CPU
{40781464-5948-462B-BA4B-98932711513F}.Release|Any CPU.Build.0 = Release|Any CPU
{3D7CF3D6-AC45-4D50-9619-5687B1443E94}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
{3D7CF3D6-AC45-4D50-9619-5687B1443E94}.Debug|Any CPU.Build.0 = Debug|Any CPU
{3D7CF3D6-AC45-4D50-9619-5687B1443E94}.Release|Any CPU.ActiveCfg = Release|Any CPU
{3D7CF3D6-AC45-4D50-9619-5687B1443E94}.Release|Any CPU.Build.0 = Release|Any CPU
EndGlobalSection
EndGlobal

42
dotnet-examples/spoken-language-identification/Program.cs Normal file
View File

@@ -0,0 +1,42 @@
// Copyright (c) 2024 Xiaomi Corporation
//
// This file shows how to do spoken language identification with whisper.
//
// 1. Download a whisper multilingual model. We use a tiny model below.
// Please refer to https://github.com/k2-fsa/sherpa-onnx/releases/tag/asr-models
// to download more models.
//
// wget https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-whisper-tiny.tar.bz2
// tar xvf sherpa-onnx-whisper-tiny.tar.bz2
// rm sherpa-onnx-whisper-tiny.tar.bz2
//
// 2. Now run it
//
// dotnet run
using SherpaOnnx;
using System.Collections.Generic;
using System;
class SpokenLanguageIdentificationDemo
{
static void Main(string[] args)
{
var config = new SpokenLanguageIdentificationConfig();
config.Whisper.Encoder = "./sherpa-onnx-whisper-tiny/tiny-encoder.int8.onnx";
config.Whisper.Decoder = "./sherpa-onnx-whisper-tiny/tiny-decoder.int8.onnx";
var slid = new SpokenLanguageIdentification(config);
var filename = "./sherpa-onnx-whisper-tiny/test_wavs/0.wav";
WaveReader waveReader = new WaveReader(filename);
var s = slid.CreateStream();
s.AcceptWaveform(waveReader.SampleRate, waveReader.Samples);
var result = slid.Compute(s);
Console.WriteLine($"Filename: {filename}");
Console.WriteLine($"Detected language: {result.Lang}");
}
}

1
dotnet-examples/spoken-language-identification/WaveReader.cs Symbolic link
View File

@@ -0,0 +1 @@
../offline-decode-files/WaveReader.cs

12
dotnet-examples/spoken-language-identification/run.sh Executable file
View File

@@ -0,0 +1,12 @@
#!/usr/bin/env bash
set -ex
if [ ! -d ./sherpa-onnx-whisper-tiny ]; then
curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-whisper-tiny.tar.bz2
tar xvf sherpa-onnx-whisper-tiny.tar.bz2
rm sherpa-onnx-whisper-tiny.tar.bz2
fi
dotnet run

15
dotnet-examples/spoken-language-identification/spoken-language-identification.csproj Normal file
View File

@@ -0,0 +1,15 @@
<Project Sdk="Microsoft.NET.Sdk">
<PropertyGroup>
<OutputType>Exe</OutputType>
<TargetFramework>net6.0</TargetFramework>
<RootNamespace>spoken_language_identification</RootNamespace>
<ImplicitUsings>enable</ImplicitUsings>
<Nullable>enable</Nullable>
</PropertyGroup>
<ItemGroup>
<PackageReference Include="org.k2fsa.sherpa.onnx" Version="*" />
</ItemGroup>
</Project>