Add live transcription progress and Docker Whisper.

Surface stage, percent, and elapsed time while jobs run, and ship Compose so local confidential transcription can use faster-whisper on port 8090.
This commit is contained in:
ben
2026-08-12 14:32:56 +02:00
parent 67c1941833
commit e0400aadef
14 changed files with 695 additions and 177 deletions
@@ -14,7 +14,7 @@ class TranscribeController extends Controller
*/
public function __invoke(TranscribeRecordingRequest $request, Recording $recording): RedirectResponse
{
if (in_array($recording->transcription_status, ['processing'], true)) {
if ($recording->transcription_status === 'processing') {
return back()->with('error', 'Transcription is already in progress.');
}
@@ -24,12 +24,16 @@ class TranscribeController extends Controller
'transcription_driver' => $driver,
'ollama_url' => $driver === 'ollama' ? rtrim($request->validated('ollama_url'), '/') : null,
'transcription_status' => 'pending',
'transcription_progress' => 'Queued — waiting to start…',
'transcription_percent' => 5,
'transcription_started_at' => now(),
'transcription_error' => null,
'transcript' => null,
'transcribed_at' => null,
]);
TranscribeRecording::dispatch($recording->fresh());
return back()->with('success', 'Transcription started. Refresh in a moment to see the result.');
return back()->with('success', 'Transcription started. Progress updates below.');
}
}
@@ -0,0 +1,17 @@
<?php
namespace App\Http\Controllers;
use App\Models\Recording;
use Illuminate\Http\JsonResponse;
class TranscriptionStatusController extends Controller
{
/**
* Live transcription progress for polling.
*/
public function __invoke(Recording $recording): JsonResponse
{
return response()->json($recording->fresh()->transcriptionStatusPayload());
}
}
+36 -7
View File
@@ -31,18 +31,32 @@ class TranscribeRecording implements ShouldQueue
*/
public function handle(TranscriptionService $transcription): void
{
$this->recording->update([
$this->recording->refresh();
$this->recording->forceFill([
'transcription_status' => 'processing',
]);
'transcription_started_at' => $this->recording->transcription_started_at ?? now(),
'transcription_error' => null,
])->save();
$this->recording->reportProgress('Preparing audio file…', 15);
try {
$text = $transcription->transcribe($this->recording);
$text = $transcription->transcribe(
$this->recording,
fn (string $message, int $percent) => $this->recording->reportProgress($message, $percent),
);
$this->recording->update([
$this->recording->reportProgress('Saving transcript…', 90);
$this->recording->forceFill([
'transcript' => $text,
'transcription_status' => 'done',
'transcription_progress' => 'Transcription complete',
'transcription_percent' => 100,
'transcription_error' => null,
'transcribed_at' => now(),
]);
])->save();
} catch (Throwable $e) {
Log::error('Transcription failed', [
'recording_id' => $this->recording->id,
@@ -50,11 +64,26 @@ class TranscribeRecording implements ShouldQueue
'message' => $e->getMessage(),
]);
$this->recording->update([
$this->recording->forceFill([
'transcription_status' => 'failed',
]);
'transcription_progress' => 'Transcription failed',
'transcription_percent' => $this->recording->transcription_percent ?: 0,
'transcription_error' => $e->getMessage(),
])->save();
throw $e;
}
}
/**
* Handle a job failure (timeouts, worker kill, etc.).
*/
public function failed(?Throwable $e): void
{
$this->recording->forceFill([
'transcription_status' => 'failed',
'transcription_progress' => 'Transcription failed',
'transcription_error' => $e?->getMessage() ?: 'Transcription stopped unexpectedly.',
])->save();
}
}
+82
View File
@@ -22,6 +22,10 @@ class Recording extends Model
'file_size_bytes',
'transcript',
'transcription_status',
'transcription_progress',
'transcription_percent',
'transcription_started_at',
'transcription_error',
'transcription_driver',
'ollama_url',
'transcribed_at',
@@ -35,8 +39,10 @@ class Recording extends Model
return [
'recorded_at' => 'datetime',
'transcribed_at' => 'datetime',
'transcription_started_at' => 'datetime',
'duration_seconds' => 'integer',
'file_size_bytes' => 'integer',
'transcription_percent' => 'integer',
];
}
@@ -57,6 +63,42 @@ class Recording extends Model
});
}
/**
* Friendly label for the selected transcription engine.
*/
protected function transcriptionDriverLabel(): Attribute
{
return Attribute::get(function (): ?string {
return match ($this->transcription_driver) {
'cloud' => 'Cloud (OpenAI Whisper)',
'local' => 'Local (faster-whisper)',
'ollama' => 'Ollama host',
default => $this->transcription_driver,
};
});
}
/**
* Whether transcription is actively running or queued.
*/
public function isTranscribing(): bool
{
return in_array($this->transcription_status, ['pending', 'processing'], true);
}
/**
* Update the live progress fields shown in the UI.
*/
public function reportProgress(string $message, int $percent, string $status = 'processing'): void
{
$this->forceFill([
'transcription_status' => $status,
'transcription_progress' => $message,
'transcription_percent' => max(0, min(100, $percent)),
'transcription_error' => null,
])->save();
}
/**
* Absolute filesystem path for the stored audio file.
*/
@@ -74,4 +116,44 @@ class Recording extends Model
Storage::disk('local')->delete($this->file_path);
}
}
/**
* Payload for the live status endpoint / Alpine poller.
*
* @return array<string, mixed>
*/
public function transcriptionStatusPayload(): array
{
$startedAt = $this->transcription_started_at;
$elapsed = $startedAt ? $startedAt->diffInSeconds(now()) : null;
return [
'id' => $this->id,
'status' => $this->transcription_status,
'progress' => $this->transcription_progress,
'percent' => $this->transcription_percent,
'driver' => $this->transcription_driver,
'driver_label' => $this->transcription_driver_label,
'error' => $this->transcription_error,
'started_at' => $startedAt?->toIso8601String(),
'elapsed_seconds' => $elapsed,
'elapsed_human' => $elapsed === null ? null : $this->formatElapsed($elapsed),
'duration_seconds' => $this->duration_seconds,
'is_active' => $this->isTranscribing(),
'has_transcript' => filled($this->transcript),
'transcribed_at' => $this->transcribed_at?->toIso8601String(),
];
}
private function formatElapsed(int $seconds): string
{
$minutes = intdiv($seconds, 60);
$remain = $seconds % 60;
if ($minutes === 0) {
return sprintf('%ds', $remain);
}
return sprintf('%dm %02ds', $minutes, $remain);
}
}
+35 -7
View File
@@ -3,6 +3,7 @@
namespace App\Services;
use App\Models\Recording;
use Closure;
use Illuminate\Support\Facades\Http;
use Illuminate\Support\Str;
use Laravel\Ai\Transcription;
@@ -12,41 +13,63 @@ class TranscriptionService
{
/**
* Run transcription for a recording using the selected driver.
*
* @param (Closure(string, int): void)|null $onProgress
*/
public function transcribe(Recording $recording): string
public function transcribe(Recording $recording, ?Closure $onProgress = null): string
{
$report = $onProgress ?? static fn (string $message, int $percent) => null;
return match ($recording->transcription_driver) {
'cloud' => $this->viaCloud($recording),
'local' => $this->viaLocal($recording),
'ollama' => $this->viaRemoteCompatible($recording),
'cloud' => $this->viaCloud($recording, $report),
'local' => $this->viaLocal($recording, $report),
'ollama' => $this->viaRemoteCompatible($recording, $report),
default => throw new RuntimeException('Unknown transcription driver: '.$recording->transcription_driver),
};
}
private function viaCloud(Recording $recording): string
/**
* @param Closure(string, int): void $report
*/
private function viaCloud(Recording $recording, Closure $report): string
{
$report('Sending audio to OpenAI Whisper…', 35);
$report('Waiting for cloud transcript (this can take a while for long recordings)…', 55);
$transcript = Transcription::fromStorage($recording->file_path)
->timeout((int) config('ai.transcription_timeout', 600))
->generate('openai', 'whisper-1');
$report('Received transcript from OpenAI…', 85);
return (string) $transcript;
}
private function viaLocal(Recording $recording): string
/**
* @param Closure(string, int): void $report
*/
private function viaLocal(Recording $recording, Closure $report): string
{
$model = config('ai.local_whisper_model', 'Systran/faster-whisper-base');
$report('Connecting to local faster-whisper server…', 30);
$report("Transcribing locally with {$model} (audio stays on this machine)…", 50);
$transcript = Transcription::fromStorage($recording->file_path)
->timeout((int) config('ai.transcription_timeout', 600))
->generate('local-whisper', $model);
$report('Received transcript from local Whisper…', 85);
return (string) $transcript;
}
/**
* Call an OpenAI-compatible /v1/audio/transcriptions endpoint at a user-supplied host URL.
*
* @param Closure(string, int): void $report
*/
private function viaRemoteCompatible(Recording $recording): string
private function viaRemoteCompatible(Recording $recording, Closure $report): string
{
if (! filled($recording->ollama_url)) {
throw new RuntimeException('Ollama host URL is required for remote transcription.');
@@ -60,6 +83,9 @@ class TranscriptionService
throw new RuntimeException('Recording audio file is not readable.');
}
$report('Connecting to remote host '.$recording->ollama_url.'…', 30);
$report("Uploading audio and waiting for transcript ({$model})…", 50);
$response = Http::timeout((int) config('ai.transcription_timeout', 600))
->attach(
'file',
@@ -83,6 +109,8 @@ class TranscriptionService
throw new RuntimeException('Remote transcription returned an empty transcript. Ensure the host exposes OpenAI-compatible /v1/audio/transcriptions.');
}
$report('Received transcript from remote host…', 85);
return $text;
}