docs: remove unwanted section

Author: tito

docs/transcript.md

@@ -239,97 +239,3 @@ All formats resolve speaker IDs to participant names when available:
 - If a participant exists for the speaker ID, their name is used
 - If no participant exists, a default name like "Speaker 0" is generated
 - Speaker IDs are integers (0, 1, 2, etc.) assigned during diarization
-
-## Edge Cases
-
-### Empty Transcripts
-
-For transcripts with no topics or empty word lists:
-
-- **text**: Returns empty string `""`
-- **text-timestamped**: Returns empty string `""`
-- **webvtt-named**: Returns minimal WebVTT header `"WEBVTT\n\n"`
-- **json**: Returns empty array `[]`
-
-### Multiple Speaker Changes
-
-All formats properly handle conversations with multiple speaker changes within a single topic, creating new segments when the speaker changes.
-
-## Examples
-
-### Using curl
-
-```bash
-# Get plain text format (default)
-curl "https://api.reflector.com/v1/transcripts/abc123"
-
-# Get timestamped text
-curl "https://api.reflector.com/v1/transcripts/abc123?transcript_format=text-timestamped"
-
-# Get WebVTT for subtitles
-curl "https://api.reflector.com/v1/transcripts/abc123?transcript_format=webvtt-named" > subtitles.vtt
-
-# Get JSON for processing
-curl "https://api.reflector.com/v1/transcripts/abc123?transcript_format=json"
-```
-
-### Using JavaScript/TypeScript
-
-```typescript
-// Using fetch API
-const response = await fetch(
-  `https://api.reflector.com/v1/transcripts/${id}?transcript_format=json`
-);
-const data = await response.json();
-
-// Type-safe access with discriminated union
-switch (data.transcript_format) {
-  case "text":
-  case "text-timestamped":
-  case "webvtt-named":
-    console.log("Text content:", data.transcript);
-    break;
-  case "json":
-    data.transcript.forEach(segment => {
-      console.log(`${segment.speaker_name}: ${segment.text}`);
-    });
-    break;
-}
-```
-
-### Using Python
-
-```python
-import requests
-
-# Get JSON format
-response = requests.get(
-    f"https://api.reflector.com/v1/transcripts/{transcript_id}",
-    params={"transcript_format": "json"}
-)
-data = response.json()
-
-# Process segments
-for segment in data["transcript"]:
-    print(f"[{segment['start']:.1f}s] {segment['speaker_name']}: {segment['text']}")
-```
-
-## Migration Notes
-
-### Backward Compatibility
-
-The GET `/v1/transcripts/{id}` endpoint previously returned transcript metadata without formatted transcript content. Adding the `transcript_format` parameter is fully backward compatible:
-
-- Without the parameter, the default `text` format is returned
-- All existing response fields remain unchanged
-- The `transcript_format` and `transcript` fields are new additions
-
-### Upgrading Client Code
-
-To use the new format feature:
-
-1. Add `transcript_format` query parameter to your requests
-2. Handle the discriminated union response type
-3. Access the `transcript` field based on the format
-
-Existing code that doesn't use these new fields will continue to work unchanged.