From b5f5ef172c836024577d161d2976fcd6d5570d73 Mon Sep 17 00:00:00 2001
From: iamdoubz <>
Date: Fri, 10 Jul 2026 19:05:28 -0500
Subject: [PATCH] feat(commands): auto-label the mic speaker "You" at
 stop_recording

Wires VoiceSample creation into start_recording (mic-enabled path) and
runs the voiceprint match after the final diarization pass in
stop_recording: on a confident match, persists "You" and "Speaker 2",
"Speaker 3"... via the existing rename_speaker store path, skipping any
label the user already renamed live.
---
 src-tauri/src/commands.rs | 59 ++++++++++++++++++++++++++++++++++++---
 1 file changed, 55 insertions(+), 4 deletions(-)

diff --git a/src-tauri/src/commands.rs b/src-tauri/src/commands.rs
index 46f6602..c2cb860 100644
--- a/src-tauri/src/commands.rs
+++ b/src-tauri/src/commands.rs
@@ -29,6 +29,11 @@ use std::path::{Path, PathBuf};
 use std::sync::{Arc, Mutex as StdMutex};
 use tauri::{AppHandle, Emitter, Manager, State};
 
+/// ~8s of 16kHz mono mic audio — enough for a stable speaker-embedding
+/// voiceprint (see `diarization::voiceprint`) without holding minutes of raw
+/// audio in memory for the whole meeting.
+const MIC_VOICEPRINT_SAMPLES: usize = 16_000 * 8;
+
 #[derive(Deserialize)]
 pub struct StartRecordingArgs {
     pub meeting_title: Option<String>,
@@ -332,11 +337,14 @@ pub async fn start_recording(
     // microphone that fails to open must not sink the meeting: we log and fall
     // back to loopback-only (the mixer forwards loopback alone once its sink
     // drops).
-    let (capture, mic_capture) = if settings.microphone_enabled {
+    let (capture, mic_capture, mic_voice_sample) = if settings.microphone_enabled {
         // The mixer sums both streams for the live transcript; the bridge carries
         // the mic into the loopback thread so the recorded WAV holds both sides
         // at native quality (FR-CAP-7).
         let bridge = crate::audio::MicBridge::shared();
+        // A few seconds of raw mic audio for the post-stop voiceprint match
+        // (bug: mic speaker mislabeled "S1"/"S2" instead of "You").
+        let voice_sample = crate::audio::VoiceSample::new(MIC_VOICEPRINT_SAMPLES);
         let (loop_sink, mic_sink) = crate::audio::spawn_mixer(frame_tx);
         let capture = WasapiCapture
             .start_loopback_recording(
@@ -353,10 +361,12 @@ pub async fn start_recording(
                 mic_sink,
                 event_tx,
                 bridge,
+                Some(voice_sample.clone()),
             )
             .map_err(|e| tracing::warn!("microphone capture unavailable: {e}"))
             .ok();
-        (capture, mic)
+        let voice_sample = mic.is_some().then_some(voice_sample);
+        (capture, mic, voice_sample)
     } else {
         let capture = WasapiCapture
             .start(
@@ -366,7 +376,7 @@ pub async fn start_recording(
                 event_tx,
             )
             .map_err(|e| WaError::new("audio", e.to_string()))?;
-        (capture, None)
+        (capture, None, None)
     };
 
     // Fire-and-forget: exits on its own once `event_tx` drops at capture stop;
@@ -554,6 +564,7 @@ pub async fn start_recording(
         language: language_state,
         diarizer,
         speaker_names,
+        mic_voice_sample,
     });
     drop(guard);
 
@@ -615,17 +626,57 @@ pub async fn stop_recording(
     // provisional passes (T4.3) produced. Skipped if diarization models
     // aren't installed — `speaker_infos_from_segments` then falls back to
     // the single pre-diarization "S1" placeholder, same as before Phase 4.
+    let mut final_spans: Option<Vec<SpeakerSpan>> = None;
     if let Some(diarizer) = session.diarizer.clone() {
         let diarizer_for_task = diarizer.clone();
         let wav_path = session.wav_path.clone();
         match tauri::async_runtime::spawn_blocking(move || diarizer_for_task.diarize(&wav_path))
             .await
         {
-            Ok(Ok(spans)) => diarizer.assign(&mut segments, &spans),
+            Ok(Ok(spans)) => {
+                diarizer.assign(&mut segments, &spans);
+                final_spans = Some(spans);
+            }
             Ok(Err(e)) => tracing::warn!("final diarization pass failed: {e}"),
             Err(e) => tracing::warn!("final diarization task failed: {e}"),
         }
     }
+
+    // Bug fix: identify which diarized cluster is the mic (voiceprint match
+    // against the mic-only sample) and auto-label it "You" — otherwise the
+    // mic speaker is just whichever cluster sherpa-onnx happened to call
+    // "S1". Never overrides a name the user already set live (T4.4).
+    if let (Some(voice_sample), Some(spans)) = (&session.mic_voice_sample, &final_spans) {
+        let mic_samples = voice_sample.samples();
+        match crate::diarization::voiceprint::match_mic_speaker(
+            &diarization_embedding_model_file(),
+            &mic_samples,
+            &session.wav_path,
+            spans,
+        ) {
+            Ok(auto_names) => {
+                for (label, name) in auto_names {
+                    let already_named = session
+                        .speaker_names
+                        .lock()
+                        .map(|g| g.contains_key(&label))
+                        .unwrap_or(true); // poisoned lock: don't guess, skip
+                    if already_named {
+                        continue;
+                    }
+                    if let Err(e) = state.store.rename_speaker(&meeting_id, &label, &name).await {
+                        tracing::warn!("failed to persist auto speaker name: {e}");
+                        continue;
+                    }
+                    if let Ok(mut names) = session.speaker_names.lock() {
+                        names.insert(label, name);
+                    }
+                }
+            }
+            Err(e) => tracing::warn!("mic voiceprint match failed: {e}"),
+        }
+    }
+
     let speaker_names = session
         .speaker_names
         .lock()