Local AI music generation removes the meter from creativity. Tools like ACE-Step 1.5 can run entirely on a Mac, turning text prompts into full songs in minutes with no cloud credits and no upload queues. That freedom breeds a new problem: volume. When generation is cheap and instant, you stop asking whether you can make a song and start wondering which of the fifty versions on your drive is worth keeping.

I learned this the hard way. On an average day, I need about four takes from ACE-Step 1.5 to find one keeper. But averages lie. In one recent session, I generated thirty-two takes chasing a single arrangement. Listening to thirty-two two-minute songs is over an hour of critical listening. By the seventh track, my ears were forgiving smeared vowels. By the fifteenth, I was nodding along to melodies while missing dropped words entirely. Listener fatigue is not laziness; it is a real collapse in perceptual accuracy. I needed a filter that could run before my ears did.

So I built a local QA pipeline around mlx-whisper, the Apple Silicon-optimized port of OpenAI’s speech-recognition model. The idea was simple: if I could transcribe every take automatically and compare it to the original lyrics, I would have an objective lyric-match rate. That number could rank the thirty-two takes down to a manageable handful.

How the Pipeline Works

The workflow has four stages, and I treat each one as non-negotiable.

Generate. I create the raw audio with ACE-Step 1.5. I do not judge anything at this stage. The goal is volume.

Transcribe. Every WAV file feeds into mlx-whisper on my Mac. Because MLX is built for Apple’s Metal and the neural engine, this runs entirely locally. There are no API costs, no network latency, and no privacy concerns about shipping raw audio to a remote server. A thirty-two-file batch transcribes while I make coffee.

Score. I compare the Whisper transcript against the original lyric prompt. The match rate measures fidelity: did the singer hit every word, or did it skip lines, slur phrases, or hallucinate syllables? I calculate a percentage overlap. This is not an aesthetic judgment; it is a strict accounting of textual accuracy.

Filter. I sort the takes by that match rate. The top quintile becomes my audition pool. Everything else goes to a secondary folder. I have not deleted the low scorers yet, but I do not waste prime listening time on them.

Keep the Jury Independent

I follow one hard rule: the generation model never scores its own homework. ACE-Step 1.5 does not evaluate ACE-Step 1.5 output. I use an entirely separate process for transcription because a model that checks itself will always be too kind. It shares the same blind spots. If the generator tends to drop plural markers or softens hard consonants, a self-evaluation loop will learn to ignore those same quirks. An independent speech-recognition model has no loyalty to the music. It simply reports what it hears, however harsh that report might be.

What the Numbers Revealed

On the batch of thirty-two takes, the pipeline narrowed the field to eight candidates that were worth serious attention. Those eight averaged an 83.9% lyric-match rate. After I listened to them properly and scored them against my own quality rubric, they averaged 94.1 out of 100. That spread tells the whole story. The machine gate caught the obvious structural failures—lyric drops, timing collapses, vocal artifacts—so that my human scoring could operate on a pre-cleaned set. The automation did not replace my judgment. It conserved it.

When the Machine Misfires

A low score is not always a bad song. I caught this early when a track I loved scored far below the cutoff. The lyrics were a simple alphabet chant: single letters sung as isolated sounds. Whisper is trained on natural sentence structure. Feed it "A B C D" and it often hallucinates words, inserts articles, or collapses the letters into garbled phonemes. The transcription failed, but the vocal performance was actually crisp.

That case taught me the practical limit. The match rate is a pre-filter, not a final verdict. Any take that scores low still gets a ten-second human audition before I trash it. The number points you toward probability, not certainty. If you treat the score as a gavel rather than a compass, you will throw away good music.

A Technical Speed Bump

Якщо ви запускаєте MLX на Apple Silicon, перевірте архітектуру вашого Python перед обробкою великих пакетів даних. Поширеною помилкою при налаштуванні є запуск бінарного файлу Python через емуляцію Rosetta. Скрипт все одно працюватиме, але ви втратите апаратне прискорення, яке робить локальну транскрипцію прийнятною.

Виконайте це у своєму терміналі:

python3 -c "import platform; print(platform.machine())"

Ви маєте побачити arm64. Якщо виводиться x86_64, ваше середовище є емульованим. Перейдіть на нативну збірку Python або нативне середовище conda, а потім перевстановіть mlx-whisper. При обробці довгих пакетів різниця між емульованим і нативним виконанням — це різниця між завершенням роботи до обіду та завершенням до вечері.

Справжня цінність

Генеративне аудіо винагороджує наполегливість, але людська увага обмежена. Немає жодної слави в тому, щоб слухати тридцять два посередні дублі лише для того, щоб довести свою ретельність. Розмістивши mlx-whisper між генератором і своїми вухами, я повернув собі години зосередженого творчого часу. Я все одно вирішую, яка пісня «живе», а яка «помирає». Машина просто гарантує, що я застосовую це судження до найкращих можливих кандидатів.

Повний опис цього пайплайну доступний тут. Якщо ви створюєте подібні локальні інструменти QA, спільнота GyaanSetu — гарне місце для обміну досвідом.