🐛 Deduplicate model names

bugbakery · Jul 7, 2023 · e0f323e · e0f323e
1 parent 3acf70b
commit e0f323e
Show file tree

Hide file tree

Showing 2 changed files with 15 additions and 5 deletions.
diff --git a/server/app/models.yml b/server/app/models.yml
@@ -15,7 +15,7 @@ German:
   size: 1.9G
   type: transcription
   compressed: true
-- name: big
+- name: big-2
   url: https://alphacephei.com/vosk/models/vosk-model-de-tuda-0.6-900k.zip
   description: Latest big wideband model from <a href="https://github.com/uhh-lt/kaldi-tuda-de">Tuda-DE</a>
     project
@@ -108,7 +108,7 @@ Russian Other:
   size: 1.5G
   type: transcription
   compressed: true
-- name: big
+- name: big-2
   url: https://alphacephei.com/vosk/models/vosk-model-ru-0.10.zip
   description: Big narrowband Russian model for servers
   size: 2.5G
@@ -235,7 +235,7 @@ Arabic:
   size: 318M
   type: transcription
   compressed: true
-- name: big
+- name: big-2
   url: https://alphacephei.com/vosk/models/vosk-model-ar-0.22-linto-1.1.0.zip
   description: Big model from <a href="https://doc.linto.ai/#/services/linstt">LINTO</a>
     project
@@ -249,7 +249,7 @@ Farsi:
   size: 47M
   type: transcription
   compressed: true
-- name: small
+- name: small-2
   url: https://alphacephei.com/vosk/models/vosk-model-small-fa-0.5.zip
   description: Bigger small model for desktop application (Persian)
   size: 60M
@@ -270,7 +270,7 @@ Ukrainian:
   size: 73M
   type: transcription
   compressed: true
-- name: small
+- name: small-2
   url: https://alphacephei.com/vosk/models/vosk-model-small-uk-v3-small.zip
   description: Small model from <a href="https://github.com/egorsmkv/speech-recognition-uk">Speech
     Recognition for Ukrainian</a>

diff --git a/server/scripts/generate_models_list.py b/server/scripts/generate_models_list.py
@@ -104,9 +104,19 @@ def print_table_from_dict_list(dict_list, columns=None):
 print_table_from_dict_list(models, columns=["lang", "name", "url"])
 
 by_language = defaultdict(list)
+names_by_language = defaultdict(set)
 for model in models:
     lang = model["lang"]
     del model["lang"]
+
+    i = 2
+    name = model["name"]
+    while name in names_by_language[lang]:
+        name = model["name"] + "-" + str(i)
+        i += 1
+    model["name"] = name
+    names_by_language[lang].add(name)
+
     by_language[lang] += [model]
 
 with open(Path(__file__).parent.parent / "app" / "models.yml", "w") as outfile: