add unknown words

.spell-check/.cspell.json

@@ -18,5 +18,6 @@
     "**/*.pv",
     "**/*.so",
     "**/*.wav",
+    "**/*.json",
   ]
 }

.spell-check/dict.txt

@@ -0,0 +1,27 @@
+Apim
+DIHARD
+Diarization
+Jaccard
+PICOVOICE
+Ryzen
+barh
+boto
+diarization
+edgecolor
+figsize
+fontsize
+jaccard
+matplotlib
+omegaconf
+picovoice
+pretrained
+protobuf
+psutil
+pvfalcon
+pyannote
+rttm
+soundfile
+tqdm
+xlim
+xticks
+ylabel

benchmark.py

@@ -199,7 +199,6 @@ def main() -> None:
     parser.add_argument("--azure-subscription-key")
     parser.add_argument("--gcp-bucket-name")
     parser.add_argument("--gcp-credentials")
-    parser.add_argument("--nemo-model-config")
     parser.add_argument("--picovoice-access-key")
     parser.add_argument("--pyannote-auth-token")
     parser.add_argument("--type", choices=[bt.value for bt in BenchmarkTypes], required=True)

dataset.py

@@ -42,6 +42,7 @@ def create(cls, x: Datasets, data_folder: str, **kwargs: Any) -> "Dataset":
 
 class VoxConverse(Dataset):
     def __init__(self, data_folder: str, label_folder: str, only_en: bool = True) -> None:
+        # / *spell - checker: disable * /
         en_audio_files = {
             "aepyx.wav", "aggyz.wav", "aiqwk.wav", "aorju.wav", "auzru.wav", "bjruf.wav", "bmsyn.wav", "bvqnu.wav",
             "bvyvm.wav", "bxcfq.wav", "cadba.wav", "cawnd.wav", "clfcg.wav", "cpebh.wav", "cqfmj.wav", "crorm.wav",
@@ -69,6 +70,7 @@ def __init__(self, data_folder: str, label_folder: str, only_en: bool = True) ->
             "ytula.wav", "yukhy.wav", "zedtj.wav", "zehzu.wav", "zowse.wav", "zqidv.wav", "zsgto.wav", "zzsba.wav",
             "zztbo.wav",
         }
+        # / *spell - checker: enable * /
         self._samples = list()
 
         files = glob.iglob(os.path.join(data_folder, "*.wav"))

requirements.txt

@@ -9,7 +9,6 @@ pvfalcon
 pyannote.audio
 pyannote.metrics
 requests
-simple-diarizer
 soundfile
 torch
 tqdm