[Misc] Benchmarks for audio models (#16505)

Signed-off-by: NickLucche <nlucches@redhat.com>
2025-04-19 11:24:14 +02:00
parent 2ef0dc53b8
commit 9d4ca19d50
4 changed files with 199 additions and 5 deletions
--- a/tests/entrypoints/openai/correctness/test_transcription_api_correctness.py
+++ b/tests/entrypoints/openai/correctness/test_transcription_api_correctness.py
@@ -150,6 +150,7 @@ def test_wer_correctness(model_name,
                         expected_wer,
                         n_examples=-1,
                         max_concurrent_request=None):
+    # TODO refactor to use `ASRDataset`
    with RemoteOpenAIServer(model_name, ['--enforce-eager']) as remote_server:
        dataset = load_hf_dataset(dataset_repo)