fix seg id error

tests/test_transcript_analysis.py

@@ -0,0 +1,68 @@
+import unittest
+from unittest.mock import MagicMock, patch
+from transcribe.strategy import TranscriptStabilityAnalyzer, TranscriptChunk, TranscriptResult, SplitMode
+
+class TestTranscriptStabilityAnalyzer(unittest.TestCase):
+    def setUp(self):
+        self.analyzer = TranscriptStabilityAnalyzer()
+
+    def test_first_chunk_yields_pending_text(self):
+        mock_chunk = MagicMock(spec=TranscriptChunk)
+        mock_chunk.join.return_value = "Hello world."
+        
+        with patch.object(self.analyzer._transcript_history, 'previous_chunk', return_value=None):
+            results = list(self.analyzer.analysis(" ", mock_chunk, buffer_duration=5.0))
+        
+        self.assertEqual(len(results), 1)
+        self.assertIsInstance(results[0], TranscriptResult)
+        self.assertIn("Hello", results[0].context)
+
+    def test_short_buffer_with_high_similarity_and_end_sentence(self):
+        curr_chunk = MagicMock(spec=TranscriptChunk)
+        curr_first = MagicMock()
+        curr_rest = [MagicMock()]
+        prev_chunk = MagicMock(spec=TranscriptChunk)
+        prev_first = MagicMock()
+
+        # Mock the items attribute
+        curr_chunk.items = [curr_first, curr_rest[0]]  # Ensure it is iterable
+        curr_chunk.get_split_first_rest.return_value = (curr_first, curr_rest)
+        prev_chunk.get_split_first_rest.return_value = (prev_first, [])
+        curr_first.compare.return_value = 0.85
+        curr_first.is_end_sentence.return_value = True
+        curr_first.has_punctuation.return_value = True
+        curr_first.join.return_value = "This is a test sentence."
+        curr_first.get_buffer_index.return_value = 0
+        curr_rest[0].join.return_value = " Continuing..."
+
+        with patch.object(self.analyzer._transcript_history, 'previous_chunk', return_value=prev_chunk):
+            with patch.object(self.analyzer._transcript_history, 'add'):
+                results = list(self.analyzer.analysis(" ", curr_chunk, buffer_duration=5.0))
+        
+        self.assertGreaterEqual(len(results), 1)
+        self.assertTrue(any(r.is_end_sentence for r in results))
+        self.assertTrue(any("test" in r.context for r in results))
+
+
+    def test_long_buffer_triggers_commit(self):
+        chunk1 = MagicMock()
+        chunk2 = MagicMock()
+        chunk3 = MagicMock()
+
+        chunk1.join.return_value = "Hello."
+        chunk2.join.return_value = "How are"
+        chunk3.join.return_value = " you?"
+
+        mock_chunk = MagicMock(spec=TranscriptChunk)
+        mock_chunk.split_by.return_value = [chunk1, chunk2, chunk3]
+        mock_chunk.get_buffer_index.return_value = 0
+
+        with patch.object(self.analyzer._transcript_history, 'previous_chunk', return_value=MagicMock()):
+            with patch.object(self.analyzer._transcript_history, 'add'):
+                results = list(self.analyzer.analysis(" ", mock_chunk, buffer_duration=15.0))
+
+        self.assertTrue(any(r.is_end_sentence for r in results))
+        self.assertTrue(any("Hello" in r.context for r in results))
+
+if __name__ == '__main__':
+    unittest.main()

transcribe/strategy.py

@@ -281,7 +281,7 @@ class TranscriptStabilityAnalyzer:
         if curr_seg_id > prev_seg_id:
             # 表示生成了一个新段落 换行
             yield TranscriptResult(
-                seg_id=prev_seg_id,
+                seg_id=curr_seg_id-1,
                 cut_index=frame_cut_index,
                 context=self._transcript_buffer.latest_paragraph,
                 is_end_sentence=True
@@ -290,7 +290,7 @@ class TranscriptStabilityAnalyzer:
         # 如果还有挂起的文本
         if (current_not_commit_text := self._transcript_buffer.current_not_commit_text.strip()):
             yield TranscriptResult(
-                seg_id=curr_seg_id+1,
+                seg_id=curr_seg_id,
                 cut_index=frame_cut_index,
                 context=current_not_commit_text
             )