Stream Whisper segments so the UI can show live transcript text.

Consume faster-whisper SSE instead of blocking on one JSON response, persist growing text and progress over Reverb, and drop the fake percent heartbeat.
2026-08-13 12:34:47 +02:00
parent 5cea5192c2
commit 4c73620458
9 changed files with 574 additions and 138 deletions
@@ -0,0 +1,91 @@
+<?php
+
+namespace Tests\Unit;
+
+use App\Services\WhisperTranscriptionStream;
+use PHPUnit\Framework\TestCase;
+
+class WhisperTranscriptionStreamTest extends TestCase
+{
+    private WhisperTranscriptionStream $stream;
+
+    protected function setUp(): void
+    {
+        parent::setUp();
+
+        $this->stream = new WhisperTranscriptionStream;
+    }
+
+    public function test_extracts_sse_data_payloads_from_buffer(): void
+    {
+        $buffer = "data: {\"type\":\"transcript.text.delta\",\"delta\":\"Hello\"}\n\npartial";
+
+        $payloads = $this->stream->extractPayloads($buffer);
+
+        $this->assertSame(['{"type":"transcript.text.delta","delta":"Hello"}'], $payloads);
+        $this->assertSame('partial', $buffer);
+    }
+
+    public function test_parses_openai_delta_and_done_events(): void
+    {
+        $delta = $this->stream->parseEvent('{"type":"transcript.text.delta","delta":"Hello "}');
+        $done = $this->stream->parseEvent('{"type":"transcript.text.done","text":"Hello world"}');
+
+        $this->assertSame('Hello ', $delta['append']);
+        $this->assertFalse($delta['done']);
+        $this->assertSame('Hello world', $done['replace']);
+        $this->assertTrue($done['done']);
+    }
+
+    public function test_parses_legacy_segment_events(): void
+    {
+        $event = $this->stream->parseEvent('{"text":"First segment","start":0,"end":12.5}');
+
+        $this->assertTrue($event['legacy']);
+        $this->assertSame('First segment', $event['append']);
+        $this->assertSame(12.5, $event['end']);
+    }
+
+    public function test_applies_delta_then_done_replace(): void
+    {
+        $delta = $this->stream->parseEvent('{"type":"transcript.text.delta","delta":"Hel"}');
+        $text = $this->stream->applyEvent('', $delta);
+        $delta2 = $this->stream->parseEvent('{"type":"transcript.text.delta","delta":"lo"}');
+        $text = $this->stream->applyEvent($text, $delta2);
+        $done = $this->stream->parseEvent('{"type":"transcript.text.done","text":"Hello world"}');
+        $text = $this->stream->applyEvent($text, $done);
+
+        $this->assertSame('Hello world', $text);
+    }
+
+    public function test_joins_legacy_segments_with_spaces(): void
+    {
+        $first = $this->stream->parseEvent('{"text":"Hello from","end":10}');
+        $second = $this->stream->parseEvent('{"text":"whisper.","end":20}');
+
+        $text = $this->stream->applyEvent('', $first);
+        $text = $this->stream->applyEvent($text, $second);
+
+        $this->assertSame('Hello from whisper.', $text);
+    }
+
+    public function test_ignores_done_sentinel_and_non_json(): void
+    {
+        $buffer = "data: [DONE]\n\ndata: not-json\n\n";
+
+        $payloads = $this->stream->extractPayloads($buffer);
+
+        $this->assertSame(['not-json'], $payloads);
+        $this->assertNull($this->stream->parseEvent('not-json'));
+    }
+
+    public function test_flush_buffer_emits_trailing_frame(): void
+    {
+        $buffer = 'data: {"text":"Last"}';
+
+        $payloads = $this->stream->flushBuffer($buffer);
+
+        $this->assertSame(['{"text":"Last"}'], $payloads);
+        $this->assertSame('', $buffer);
+    }
+}