meeting-assistant/MeetingAssistant/Recording/SpeakerAudioSampleCollector.cs

using MeetingAssistant.Speakers;
using MeetingAssistant.Transcription;

namespace MeetingAssistant.Recording;

internal sealed class SpeakerAudioSampleCollector
{
    private readonly object gate = new();
    private readonly RollingAudioBuffer audioBuffer;
    private readonly Dictionary<string, List<SpeakerAudioSample>> samplesBySpeaker = new(StringComparer.OrdinalIgnoreCase);
    private readonly int maxSamplesPerSpeaker;

    public SpeakerAudioSampleCollector(TimeSpan bufferDuration, int maxSamplesPerSpeaker)
    {
        audioBuffer = new RollingAudioBuffer(bufferDuration);
        this.maxSamplesPerSpeaker = Math.Max(1, maxSamplesPerSpeaker);
    }

    public void AppendAudio(AudioChunk chunk)
    {
        audioBuffer.Append(chunk);
    }

    public void Reset()
    {
        lock (gate)
        {
            samplesBySpeaker.Clear();
            audioBuffer.Reset();
        }
    }

    public void TryAdd(TranscriptionSegment segment)
    {
        if (!IsDiarizedSpeaker(segment.Speaker))
        {
            return;
        }

        var score = Score(segment);
        if (score <= 0)
        {
            return;
        }

        var wavBytes = audioBuffer.TryExtractWav(segment.Start, segment.End);
        if (wavBytes.Length == 0)
        {
            return;
        }

        var sample = new SpeakerAudioSample(segment.Speaker, segment, wavBytes, score);
        lock (gate)
        {
            if (!samplesBySpeaker.TryGetValue(segment.Speaker, out var samples))
            {
                samples = [];
                samplesBySpeaker[segment.Speaker] = samples;
            }

            samples.Add(sample);
            var bestSamples = samples
                .OrderByDescending(candidate => candidate.Score)
                .Take(maxSamplesPerSpeaker)
                .ToList();
            samples.Clear();
            samples.AddRange(bestSamples);
        }
    }

    public IReadOnlyList<SpeakerAudioSample> Snapshot()
    {
        lock (gate)
        {
            return samplesBySpeaker.Values
                .SelectMany(samples => samples)
                .OrderBy(sample => sample.Speaker, StringComparer.OrdinalIgnoreCase)
                .ThenByDescending(sample => sample.Score)
                .ToList();
        }
    }

    private static bool IsDiarizedSpeaker(string speaker)
    {
        return !string.IsNullOrWhiteSpace(speaker) &&
            !string.Equals(speaker, "Unknown", StringComparison.OrdinalIgnoreCase);
    }

    private static double Score(TranscriptionSegment segment)
    {
        var durationSeconds = (segment.End - segment.Start).TotalSeconds;
        if (durationSeconds < 2 || durationSeconds > 30)
        {
            return 0;
        }

        var words = segment.Text
            .Split(' ', StringSplitOptions.RemoveEmptyEntries | StringSplitOptions.TrimEntries)
            .Length;
        if (words < 3 || words > 80)
        {
            return 0;
        }

        var durationScore = 1 - Math.Min(Math.Abs(durationSeconds - 8) / 8, 1);
        var wordScore = 1 - Math.Min(Math.Abs(words - 18) / 18.0, 1);
        var sentenceBonus = segment.Text.TrimEnd().EndsWith('.') ||
            segment.Text.TrimEnd().EndsWith('?') ||
            segment.Text.TrimEnd().EndsWith('!')
                ? 5
                : 0;
        return durationScore * 70 + wordScore * 30 + sentenceBonus;
    }
}