Add meeting assistant speech and summary automation

2026-05-21 09:55:39 +02:00
parent 1e97d2b9ff
commit 48d98d9cbb
63 changed files with 5143 additions and 151 deletions
@@ -0,0 +1,148 @@
+using MeetingAssistant;
+using MeetingAssistant.Speakers;
+using MeetingAssistant.Transcription;
+using Microsoft.Extensions.Logging.Abstractions;
+using Microsoft.Extensions.Options;
+using NAudio.Wave;
+
+namespace MeetingAssistant.Tests;
+
+public sealed class AzureSpeechSpeakerIdentityMatcherTests
+{
+    [Fact]
+    public void IdentityDiarizationLanguageUsesFirstAutoDetectLanguageWhenConfiguredLanguageIsAuto()
+    {
+        Assert.Equal(
+            "de-DE",
+            AzureSpeechSpeakerIdentityDiarizationClient.ResolveRecognitionLanguage(new AzureSpeechOptions
+            {
+                Language = "auto",
+                AutoDetectLanguages = ["de-DE", "en-US"]
+            }));
+    }
+
+    [Fact]
+    public void IdentityDiarizationLanguageUsesConfiguredConcreteLanguage()
+    {
+        Assert.Equal(
+            "en-US",
+            AzureSpeechSpeakerIdentityDiarizationClient.ResolveRecognitionLanguage(new AzureSpeechOptions
+            {
+                Language = "en-US",
+                AutoDetectLanguages = ["de-DE"]
+            }));
+    }
+
+    [Fact]
+    public async Task MatcherReturnsIdentityWhenAzureDiarizationAssignsUnknownSnippetToKnownSpeaker()
+    {
+        var diarizationClient = new FakeSpeakerIdentityDiarizationClient(
+        [
+            new TranscriptionSegment(TimeSpan.Zero, TimeSpan.FromSeconds(1), "Guest-1", "known"),
+            new TranscriptionSegment(TimeSpan.FromSeconds(2), TimeSpan.FromSeconds(3), "Guest-1", "unknown")
+        ]);
+        var matcher = CreateMatcher(diarizationClient);
+        var snippet = CreateWavSnippet();
+
+        var match = await matcher.MatchAsync(
+            new SpeakerIdentityMatchRequest(
+                "Guest03",
+                snippet,
+                [new SpeakerIdentityMatchCandidate(42, "Chris", 5, [snippet])]),
+            CancellationToken.None);
+
+        Assert.NotNull(match);
+        Assert.Equal(42, match.IdentityId);
+        Assert.NotNull(diarizationClient.WavPath);
+        Assert.False(File.Exists(diarizationClient.WavPath));
+    }
+
+    [Fact]
+    public async Task MatcherRequiresTwoMatchingKnownSnippetsWhenIdentityHasMultipleSnippets()
+    {
+        var diarizationClient = new FakeSpeakerIdentityDiarizationClient(
+        [
+            new TranscriptionSegment(TimeSpan.Zero, TimeSpan.FromSeconds(1), "Guest-1", "known one"),
+            new TranscriptionSegment(TimeSpan.FromSeconds(2), TimeSpan.FromSeconds(3), "Guest-2", "known two"),
+            new TranscriptionSegment(TimeSpan.FromSeconds(4), TimeSpan.FromSeconds(5), "Guest-1", "unknown")
+        ]);
+        var matcher = CreateMatcher(diarizationClient);
+        var snippet = CreateWavSnippet();
+
+        var match = await matcher.MatchAsync(
+            new SpeakerIdentityMatchRequest(
+                "Guest03",
+                snippet,
+                [new SpeakerIdentityMatchCandidate(42, "Chris", 5, [snippet, snippet])]),
+            CancellationToken.None);
+
+        Assert.Null(match);
+    }
+
+    [Fact]
+    public async Task MatcherReturnsNullWhenAzureDiarizationAssignsDifferentSpeakers()
+    {
+        var diarizationClient = new FakeSpeakerIdentityDiarizationClient(
+        [
+            new TranscriptionSegment(TimeSpan.Zero, TimeSpan.FromSeconds(1), "Guest-1", "known"),
+            new TranscriptionSegment(TimeSpan.FromSeconds(2), TimeSpan.FromSeconds(3), "Guest-2", "unknown")
+        ]);
+        var matcher = CreateMatcher(diarizationClient);
+        var snippet = CreateWavSnippet();
+
+        var match = await matcher.MatchAsync(
+            new SpeakerIdentityMatchRequest(
+                "Guest03",
+                snippet,
+                [new SpeakerIdentityMatchCandidate(42, "Chris", 5, [snippet])]),
+            CancellationToken.None);
+
+        Assert.Null(match);
+    }
+
+    private static AzureSpeechSpeakerIdentityMatcher CreateMatcher(ISpeakerIdentityDiarizationClient diarizationClient)
+    {
+        return new AzureSpeechSpeakerIdentityMatcher(
+            diarizationClient,
+            Options.Create(new MeetingAssistantOptions
+            {
+                SpeakerIdentification = new SpeakerIdentificationOptions
+                {
+                    SilenceBetweenSnippetsSeconds = 1
+                }
+            }),
+            NullLogger<AzureSpeechSpeakerIdentityMatcher>.Instance);
+    }
+
+    private static byte[] CreateWavSnippet()
+    {
+        using var stream = new MemoryStream();
+        using (var writer = new WaveFileWriter(stream, new WaveFormat(16000, 16, 1)))
+        {
+            writer.Write(new byte[16000 * 2], 0, 16000 * 2);
+        }
+
+        return stream.ToArray();
+    }
+
+    private sealed class FakeSpeakerIdentityDiarizationClient : ISpeakerIdentityDiarizationClient
+    {
+        private readonly IReadOnlyList<TranscriptionSegment> segments;
+
+        public FakeSpeakerIdentityDiarizationClient(IReadOnlyList<TranscriptionSegment> segments)
+        {
+            this.segments = segments;
+        }
+
+        public string? WavPath { get; private set; }
+
+        public Task<IReadOnlyList<TranscriptionSegment>> DiarizeAsync(
+            string wavPath,
+            CancellationToken cancellationToken)
+        {
+            WavPath = wavPath;
+            Assert.True(File.Exists(wavPath));
+            return Task.FromResult(segments);
+        }
+    }
+}