Skip to content

Add missing integration tests for ISpeechToTextClient. #377

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
sandrohanea merged 10 commits into from
Apr 10, 2025
Merged

Add missing integration tests for ISpeechToTextClient. #377

Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
c998e76
WIP SpeechToText Impl
rogerbarreto Apr 6, 2025
212436b
Extensions Added
rogerbarreto Apr 6, 2025
e7a6285
Update impl for MEAI SpeechToText
rogerbarreto Apr 6, 2025
896b992
Update to released package
rogerbarreto Apr 9, 2025
0a579a3
Minor updates
rogerbarreto Apr 9, 2025
f7ec811
Add missing Unit Tests
rogerbarreto Apr 10, 2025
69a8bc5
Adding factory + process ctor + tests
rogerbarreto Apr 10, 2025
62c5d17
Ensure disposal works as expected
rogerbarreto Apr 10, 2025
d0e05ff
Add missing IT
rogerbarreto Apr 10, 2025
8d0e53d
Merge branch 'main' of https://github.com/sandrohanea/whisper.net int…
rogerbarreto Apr 10, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
92 changes: 49 additions & 43 deletions tests/Whisper.net.Tests/SpeechToText/WhisperSpeechToTextClientTest.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,101 +9,106 @@
namespace Whisper.net.Tests;
public partial class WhisperSpeechToTextClientTest(TinyModelFixture model) : IClassFixture<TinyModelFixture>
{
/*
[Fact]
public async Task TestHappyFlowAsync()
{
var segments = new List<SegmentData>();
var segmentsEnumerated = new List<SegmentData>();
var updatesEnumerated = new List<SpeechToTextResponseUpdate>();
var progress = new List<int>();

var encoderBegins = new List<EncoderBeginData>();
using var factory = WhisperFactory.FromPath(model.ModelFile);
using var processor = factory.CreateBuilder()

var options = new SpeechToTextOptions()
.WithLanguage("en")
.WithEncoderBeginHandler((e) =>
{
encoderBegins.Add(e);
return true;
})
.WithProgressHandler(progress.Add)
.WithSegmentEventHandler(segments.Add)
.Build();
.WithSegmentEventHandler(segments.Add);

using var client = new WhisperSpeechToTextClient(() => factory);

using var fileReader = await TestDataProvider.OpenFileStreamAsync("kennedy.wav");
await foreach (var data in processor.ProcessAsync(fileReader))
await foreach (var data in client.GetStreamingTextAsync(fileReader, options))
{
segmentsEnumerated.Add(data);
updatesEnumerated.Add(data);
}

Assert.Equal(segments, segmentsEnumerated);
Assert.Equal(segments, updatesEnumerated.Select(u => u.RawRepresentation));
Assert.True(segments.Count > 0);
Assert.True(progress.SequenceEqual(progress.OrderBy(x => x)));
Assert.True(progress.Count > 1);
Assert.Single(encoderBegins);
Assert.Contains(segments, segmentData => segmentData.Text.Contains("nation should commit"));
Assert.Contains(updatesEnumerated, update => update.Text.Contains("nation should commit"));
}

[Fact]
public async Task ProcessAsync_Cancelled_WillCancellTheProcessing_AndDispose_WillWaitUntilFullyFinished()
public async Task WithSegmentEventHandler_Cancelled_WillCancellTheProcessing_AndDispose()
{
var segments = new List<SegmentData>();
var segmentsEnumerated = new List<SegmentData>();
var segmentsEnumerated = new List<SpeechToTextResponseUpdate>();
var cts = new CancellationTokenSource();
TaskCanceledException? taskCanceledException = null;

var encoderBegins = new List<EncoderBeginData>();
using var factory = WhisperFactory.FromPath(model.ModelFile);
var processor = factory.CreateBuilder()
.WithLanguage("en")
.WithEncoderBeginHandler((e) =>
{
encoderBegins.Add(e);
return true;
})
.WithSegmentEventHandler(s =>
{
segments.Add(s);
cts.Cancel();
})
.Build();

var options = new SpeechToTextOptions()
.WithLanguage("en")
.WithEncoderBeginHandler((e) =>
{
encoderBegins.Add(e);
return true;
})
.WithSegmentEventHandler(s =>
{
segments.Add(s);
cts.Cancel();
});

var client = new WhisperSpeechToTextClient(() => factory);

using var fileReader = await TestDataProvider.OpenFileStreamAsync("kennedy.wav");
try
{
await foreach (var data in processor.ProcessAsync(fileReader, cts.Token))
await foreach (var update in client.GetStreamingTextAsync(fileReader, options, cts.Token))
{
segmentsEnumerated.Add(data);
segmentsEnumerated.Add(update);
}
}
catch (TaskCanceledException ex)
{
taskCanceledException = ex;
}

await processor.DisposeAsync();
client.Dispose();

Assert.Empty(segmentsEnumerated);
Assert.Single( segments);
Assert.Single( encoderBegins);
Assert.Single(segments);
Assert.Single(encoderBegins);
Assert.NotNull(taskCanceledException);
Assert.Contains(segments, segmentData => segmentData.Text.Contains("nation should commit"));
}

[Fact]
public async Task ProcessAsync_WhenJunkChunkExists_ProcessCorrectly()
{
var segments = new List<SegmentData>();
var segments = new List<SpeechToTextResponseUpdate>();

using var factory = WhisperFactory.FromPath(model.ModelFile);
await using var processor = factory.CreateBuilder()
.WithLanguage("en")
.Build();
var options = new SpeechToTextOptions()
.WithLanguage("en");

using var client = new WhisperSpeechToTextClient(() => factory);

using var fileReader = await TestDataProvider.OpenFileStreamAsync("junkchunk16khz.wav");
await foreach (var segment in processor.ProcessAsync(fileReader))
await foreach (var update in client.GetStreamingTextAsync(fileReader, options))
{
segments.Add(segment);
segments.Add(update);
}

Assert.True(segments.Count >= 1);
Expand All @@ -112,21 +117,22 @@ public async Task ProcessAsync_WhenJunkChunkExists_ProcessCorrectly()
[Fact]
public async Task ProcessAsync_WhenMultichannel_ProcessCorrectly()
{
var segments = new List<SegmentData>();
var segments = new List<SpeechToTextResponseUpdate>();

using var factory = WhisperFactory.FromPath(model.ModelFile);
await using var processor = factory.CreateBuilder()
.WithLanguage("en")
.Build();
var options = new SpeechToTextOptions()
.WithLanguage("en");

using var client = new WhisperSpeechToTextClient(() => factory);

using var fileReader = await TestDataProvider.OpenFileStreamAsync("multichannel.wav");
await foreach (var segment in processor.ProcessAsync(fileReader))
await foreach (var update in client.GetStreamingTextAsync(fileReader, options))
{
segments.Add(segment);
segments.Add(update);
}

Assert.True(segments.Count >= 1);
}*/
}

[Fact]
public async Task GetStreamingTextAsync_CalledMultipleTimes_Serially_WillCompleteEverytime()
Expand All @@ -135,7 +141,7 @@ public async Task GetStreamingTextAsync_CalledMultipleTimes_Serially_WillComplet
var updates2 = new List<SpeechToTextResponseUpdate>();
var updates3 = new List<SpeechToTextResponseUpdate>();

var client = new WhisperSpeechToTextClient(model.ModelFile);
using var client = new WhisperSpeechToTextClient(model.ModelFile);
var options = new SpeechToTextOptions().WithLanguage("en");

using var fileReader = await TestDataProvider.OpenFileStreamAsync("kennedy.wav");
Expand Down Expand Up @@ -163,7 +169,7 @@ public async Task GetStreamingTextAsync_CalledMultipleTimes_Serially_WillComplet
[Fact]
public async Task GetTextAsync_CalledMultipleTimes_Serially_WillCompleteEverytime()
{
var client = new WhisperSpeechToTextClient(model.ModelFile);
using var client = new WhisperSpeechToTextClient(model.ModelFile);
var options = new SpeechToTextOptions().WithLanguage("en");

using var fileReader1 = await TestDataProvider.OpenFileStreamAsync("kennedy.wav");
Expand Down