Files
meeting-assistant/MeetingAssistant/Speakers/SpeakerIdentityService.cs
T

380 lines
15 KiB
C#

using MeetingAssistant.Transcription;
using Microsoft.EntityFrameworkCore;
using Microsoft.Extensions.Options;
namespace MeetingAssistant.Speakers;
public sealed class SpeakerIdentityService : ISpeakerIdentificationService
{
private readonly IDbContextFactory<SpeakerIdentityDbContext> dbContextFactory;
private readonly ISpeakerSnippetExtractor snippetExtractor;
private readonly ISpeakerIdentityMatcher matcher;
private readonly SpeakerIdentificationOptions options;
private readonly ILogger<SpeakerIdentityService> logger;
public SpeakerIdentityService(
IDbContextFactory<SpeakerIdentityDbContext> dbContextFactory,
ISpeakerSnippetExtractor snippetExtractor,
ISpeakerIdentityMatcher matcher,
IOptions<MeetingAssistantOptions> options,
ILogger<SpeakerIdentityService> logger)
{
this.dbContextFactory = dbContextFactory;
this.snippetExtractor = snippetExtractor;
this.matcher = matcher;
this.options = options.Value.SpeakerIdentification;
this.logger = logger;
}
public async Task<SpeakerIdentificationResult> ProcessFinishedTranscriptAsync(
SpeakerIdentificationRequest request,
CancellationToken cancellationToken)
{
return await ProcessTranscriptAsync(request, mode: SpeakerIdentityProcessingMode.Final, cancellationToken);
}
public async Task<SpeakerIdentificationResult> IdentifyKnownSpeakersAsync(
SpeakerIdentificationRequest request,
CancellationToken cancellationToken)
{
return await ProcessTranscriptAsync(request, mode: SpeakerIdentityProcessingMode.LiveReadOnly, cancellationToken);
}
private async Task<SpeakerIdentificationResult> ProcessTranscriptAsync(
SpeakerIdentificationRequest request,
SpeakerIdentityProcessingMode mode,
CancellationToken cancellationToken)
{
if (!options.Enabled || request.Segments.Count == 0)
{
return new SpeakerIdentificationResult(request.Segments, new Dictionary<string, string>());
}
await using var context = await dbContextFactory.CreateDbContextAsync(cancellationToken);
await SpeakerIdentitySchema.EnsureCreatedOrUpdatedAsync(context, cancellationToken);
var attendees = NormalizeAttendees(request.MeetingNote.Frontmatter.Attendees);
var speakerMappings = new Dictionary<string, string>(StringComparer.OrdinalIgnoreCase);
var attendeeMatches = new List<SpeakerIdentityAttendeeMatch>();
var matchedAcceptedNames = new HashSet<string>(StringComparer.OrdinalIgnoreCase);
var unmatchedSpeakers = new List<(string Speaker, byte[] Snippet)>();
var samplesBySpeaker = request.Samples?
.Where(sample => !string.IsNullOrWhiteSpace(sample.Speaker) && sample.WavBytes.Length > 0)
.GroupBy(sample => sample.Speaker, StringComparer.OrdinalIgnoreCase)
.ToDictionary(
group => group.Key,
group => group.OrderByDescending(sample => sample.Score).Select(sample => sample.WavBytes).First(),
StringComparer.OrdinalIgnoreCase)
?? new Dictionary<string, byte[]>(StringComparer.OrdinalIgnoreCase);
foreach (var group in request.Segments
.Where(segment => !string.IsNullOrWhiteSpace(segment.Speaker))
.GroupBy(segment => segment.Speaker)
.OrderBy(group => group.Min(segment => segment.Start)))
{
var speaker = group.Key;
if (speakerMappings.ContainsKey(speaker))
{
continue;
}
var snippet = samplesBySpeaker.TryGetValue(speaker, out var suppliedSnippet)
? suppliedSnippet
: await snippetExtractor.ExtractSnippetAsync(
request.AudioPath,
group.ToList(),
cancellationToken);
if (snippet.Length == 0)
{
unmatchedSpeakers.Add((speaker, snippet));
continue;
}
var match = await FindMatchAsync(context, attendees, speaker, snippet, cancellationToken);
if (match is null)
{
unmatchedSpeakers.Add((speaker, snippet));
continue;
}
var identity = await LoadIdentityAsync(context, match.IdentityId, cancellationToken);
if (identity is null)
{
unmatchedSpeakers.Add((speaker, snippet));
continue;
}
if (mode == SpeakerIdentityProcessingMode.Final)
{
UpdateMatchedIdentity(identity, attendees, snippet);
foreach (var acceptedName in GetAcceptedNames(identity))
{
matchedAcceptedNames.Add(acceptedName);
}
}
var speakerName = identity.GetDisplayName();
if (!string.IsNullOrWhiteSpace(speakerName))
{
speakerMappings[speaker] = speakerName;
attendeeMatches.Add(new SpeakerIdentityAttendeeMatch(
speakerName,
GetAcceptedNames(identity).ToList()));
}
}
if (mode == SpeakerIdentityProcessingMode.Final)
{
await LearnUnmatchedSpeakersAsync(context, attendees, matchedAcceptedNames, unmatchedSpeakers, cancellationToken);
}
await context.SaveChangesAsync(cancellationToken);
var relabeledSegments = request.Segments
.Select(segment => speakerMappings.TryGetValue(segment.Speaker, out var speakerName)
? segment with { Speaker = speakerName }
: segment)
.ToList();
return new SpeakerIdentificationResult(relabeledSegments, speakerMappings, attendeeMatches);
}
private async Task<SpeakerIdentityMatch?> FindMatchAsync(
SpeakerIdentityDbContext context,
IReadOnlyList<string> attendees,
string speaker,
byte[] snippet,
CancellationToken cancellationToken)
{
var activeCutoff = DateTimeOffset.UtcNow - options.MatchIdentityActiveAge;
var maxCandidates = Math.Max(1, options.MaxMatchCandidates);
var identities = await context.SpeakerIdentities
.Include(identity => identity.Snippets)
.Include(identity => identity.Aliases)
.OrderByDescending(identity => identity.TranscriptionCount)
.ThenBy(identity => identity.Id)
.ToListAsync(cancellationToken);
identities = identities
.Select(identity => new
{
Identity = identity,
IsAttendee = MatchesAttendees(identity, attendees),
IsActive = identity.UpdatedAt >= activeCutoff
})
.Where(candidate => candidate.IsAttendee || candidate.IsActive)
.OrderByDescending(candidate => candidate.IsAttendee)
.ThenByDescending(candidate => candidate.Identity.TranscriptionCount)
.ThenBy(candidate => candidate.Identity.Id)
.Take(maxCandidates)
.Select(candidate => candidate.Identity)
.ToList();
foreach (var batch in identities.Chunk(Math.Max(1, options.MatchBatchSize)))
{
var request = new SpeakerIdentityMatchRequest(
speaker,
snippet,
batch.Select(identity => new SpeakerIdentityMatchCandidate(
identity.Id,
identity.CanonicalName,
identity.TranscriptionCount,
identity.Snippets.Select(storedSnippet => storedSnippet.WavBytes).ToList()))
.ToList());
var match = await matcher.MatchAsync(request, cancellationToken);
if (match is not null)
{
return match;
}
}
return null;
}
private static bool MatchesAttendees(
SpeakerIdentity identity,
IReadOnlyList<string> attendees)
{
var attendeeSet = attendees.ToHashSet(StringComparer.OrdinalIgnoreCase);
return GetAcceptedNames(identity).Any(attendeeSet.Contains);
}
private static Task<SpeakerIdentity?> LoadIdentityAsync(
SpeakerIdentityDbContext context,
int identityId,
CancellationToken cancellationToken)
{
return context.SpeakerIdentities
.Include(identity => identity.CandidateNames)
.Include(identity => identity.Aliases)
.Include(identity => identity.Snippets)
.SingleOrDefaultAsync(identity => identity.Id == identityId, cancellationToken);
}
private void UpdateMatchedIdentity(
SpeakerIdentity identity,
IReadOnlyList<string> attendees,
byte[] snippet)
{
identity.TranscriptionCount++;
identity.UpdatedAt = DateTimeOffset.UtcNow;
if (string.IsNullOrWhiteSpace(identity.CanonicalName) && attendees.Count > 0)
{
var currentCandidates = identity.CandidateNames
.Select(candidate => candidate.Name)
.ToHashSet(StringComparer.OrdinalIgnoreCase);
var fallbackAliasCandidate = currentCandidates.Count == 1 ? currentCandidates.Single() : null;
var aliasToCandidate = identity.Aliases
.Where(alias => !string.IsNullOrWhiteSpace(alias.Name))
.SelectMany(alias => currentCandidates.Select(candidate => new { Alias = alias.Name, Candidate = candidate }))
.Where(pair => string.Equals(pair.Alias, pair.Candidate, StringComparison.OrdinalIgnoreCase) ||
pair.Alias.Contains(pair.Candidate, StringComparison.OrdinalIgnoreCase) ||
pair.Candidate.Contains(pair.Alias, StringComparison.OrdinalIgnoreCase))
.ToDictionary(pair => pair.Alias, pair => pair.Candidate, StringComparer.OrdinalIgnoreCase);
var intersection = attendees
.Select(attendee => currentCandidates.Contains(attendee)
? attendee
: aliasToCandidate.GetValueOrDefault(attendee) ??
(identity.Aliases.Any(alias => string.Equals(alias.Name, attendee, StringComparison.OrdinalIgnoreCase))
? fallbackAliasCandidate
: null))
.Where(candidate => !string.IsNullOrWhiteSpace(candidate))
.Select(candidate => candidate!)
.Distinct(StringComparer.OrdinalIgnoreCase)
.Order(StringComparer.OrdinalIgnoreCase)
.ToList();
if (intersection.Count == 0)
{
ResetCandidates(identity, attendees);
ReplaceOldestSnippet(identity, snippet);
return;
}
ResetCandidates(identity, intersection);
if (intersection.Count == 1)
{
identity.CanonicalName = intersection[0];
}
}
AddSnippetIfNeeded(identity, snippet);
}
private async Task LearnUnmatchedSpeakersAsync(
SpeakerIdentityDbContext context,
IReadOnlyList<string> attendees,
IEnumerable<string> matchedCanonicalNames,
IReadOnlyList<(string Speaker, byte[] Snippet)> unmatchedSpeakers,
CancellationToken cancellationToken)
{
var remainingCandidates = attendees
.Except(matchedCanonicalNames, StringComparer.OrdinalIgnoreCase)
.Order(StringComparer.OrdinalIgnoreCase)
.ToList();
if (remainingCandidates.Count == 0)
{
return;
}
foreach (var (_, snippet) in unmatchedSpeakers.Where(speaker => speaker.Snippet.Length > 0))
{
var now = DateTimeOffset.UtcNow;
context.SpeakerIdentities.Add(new SpeakerIdentity
{
CreatedAt = now,
UpdatedAt = now,
CandidateNames = remainingCandidates
.Select(candidate => new SpeakerCandidateName { Name = candidate })
.ToList(),
Snippets =
[
new SpeakerSnippet
{
WavBytes = snippet,
CreatedAt = now
}
]
});
}
await Task.CompletedTask;
}
private void AddSnippetIfNeeded(SpeakerIdentity identity, byte[] snippet)
{
if (snippet.Length == 0 || identity.Snippets.Count >= options.MaxSnippetsPerSpeaker)
{
return;
}
identity.Snippets.Add(new SpeakerSnippet
{
WavBytes = snippet,
CreatedAt = DateTimeOffset.UtcNow
});
}
private static void ReplaceOldestSnippet(SpeakerIdentity identity, byte[] snippet)
{
var oldest = identity.Snippets.OrderBy(storedSnippet => storedSnippet.CreatedAt).FirstOrDefault();
if (oldest is not null)
{
identity.Snippets.Remove(oldest);
}
if (snippet.Length > 0)
{
identity.Snippets.Add(new SpeakerSnippet
{
WavBytes = snippet,
CreatedAt = DateTimeOffset.UtcNow
});
}
}
private static void ResetCandidates(SpeakerIdentity identity, IReadOnlyList<string> candidates)
{
identity.CandidateNames.Clear();
identity.CandidateNames.AddRange(candidates
.Distinct(StringComparer.OrdinalIgnoreCase)
.Order(StringComparer.OrdinalIgnoreCase)
.Select(candidate => new SpeakerCandidateName { Name = candidate }));
}
private static IReadOnlySet<string> GetAcceptedNames(SpeakerIdentity identity)
{
return new[]
{
identity.CanonicalName
}
.Concat(identity.Aliases.Select(alias => alias.Name))
.Concat(identity.CandidateNames.Select(candidate => candidate.Name))
.Where(name => !string.IsNullOrWhiteSpace(name))
.Select(name => name!.Trim())
.ToHashSet(StringComparer.OrdinalIgnoreCase);
}
private static IReadOnlyList<string> NormalizeAttendees(IEnumerable<string> attendees)
{
return attendees
.Select(NormalizeAttendee)
.Where(attendee => !string.IsNullOrWhiteSpace(attendee))
.Distinct(StringComparer.OrdinalIgnoreCase)
.Order(StringComparer.OrdinalIgnoreCase)
.ToList();
}
private static string NormalizeAttendee(string attendee)
{
var trimmed = attendee.Trim();
var emailStart = trimmed.IndexOf('<', StringComparison.Ordinal);
return emailStart > 0 ? trimmed[..emailStart].Trim() : trimmed;
}
private enum SpeakerIdentityProcessingMode
{
LiveReadOnly,
Final
}
}