fix(languages): map English (India) to Whisper code en (#617)

* fix(languages): map en-in to Whisper code en

English (India) used the catalog id as the Whisper language argument,
which is invalid. Resolve Whisper/whisper.cpp/remote_api codes via the
SUPPORTED_LANGUAGES whisper field so en-us and en-in both become en.

Co-authored-by: Jatin K Malik <jatinkrmalik@users.noreply.github.com>

* test(languages): cover whisper language resolver fallbacks

Exercise unknown catalog ids and legacy en-us fallback paths so
codecov/patch covers resolve_whisper_language fully.

Co-authored-by: Jatin K Malik <jatinkrmalik@users.noreply.github.com>

---------

Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: Jatin K Malik <jatinkrmalik@users.noreply.github.com>

src/vocalinux/speech_recognition/recognition_manager.py

@@ -21,13 +21,31 @@ from typing import Callable, Optional
 from ..common_types import RecognitionState
 from ..ui.audio_feedback import play_error_sound, play_start_sound, play_stop_sound
 from ..utils.paths import models_dir
-from ..utils.vosk_model_info import VOSK_MODEL_INFO
+from ..utils.vosk_model_info import SUPPORTED_LANGUAGES, VOSK_MODEL_INFO
 from ..utils.whispercpp_model_info import WHISPERCPP_MODEL_INFO, get_model_path, is_model_downloaded
 from ..version import __version__
 from .command_processor import CommandProcessor
 from .silero_vad import SILERO_CHUNK_SIZE, load_silero_vad
 
 
+def resolve_whisper_language(language: str) -> Optional[str]:
+    """Map a catalog language id to a Whisper / whisper.cpp language code.
+
+    Catalog keys like ``en-us`` / ``en-in`` are not valid Whisper codes; the
+    ``whisper`` field on ``SUPPORTED_LANGUAGES`` holds the ISO-style code
+    (``en``). ``auto`` resolves to ``None`` for model auto-detect.
+    """
+    if language == "auto":
+        return None
+    info = SUPPORTED_LANGUAGES.get(language)
+    if info is not None and "whisper" in info:
+        return info["whisper"]
+    # Legacy fallback for codes not present in the catalog.
+    if language == "en-us":
+        return "en"
+    return language
+
+
 # ALSA error handler to suppress warnings during PyAudio initialization
 def _setup_alsa_error_handler():
     """Set up an error handler to suppress ALSA warnings."""
@@ -1137,11 +1155,7 @@ class SpeechRecognitionManager:
                     import torch
                 use_fp16 = self.model.device != torch.device("cpu")
 
-                lang = self.language
-                if self.language == "en-us":
-                    lang = "en"
-                elif self.language == "auto":
-                    lang = None  # Auto-detect
+                lang = resolve_whisper_language(self.language)
 
                 # Transcribe with Whisper (handles variable length audio automatically)
                 result = self.model.transcribe(
@@ -1488,11 +1502,7 @@ class SpeechRecognitionManager:
             )
 
             # Prepare language parameter
-            lang = self.language
-            if self.language == "en-us":
-                lang = "en"
-            elif self.language == "auto":
-                lang = None  # Auto-detect
+            lang = resolve_whisper_language(self.language)
 
             logger.debug(f"whisper.cpp using language: {lang or 'auto-detect'}")
 
@@ -1625,11 +1635,7 @@ class SpeechRecognitionManager:
             )
 
             # Prepare language parameters
-            lang = self.language
-            if lang == "en-us":
-                lang = "en"
-            elif lang == "auto":
-                lang = None
+            lang = resolve_whisper_language(self.language)
 
             # Prepare HTTP request headers
             headers = {}

tests/test_recognition_manager_remote_api.py

@@ -366,6 +366,18 @@ class TestRemoteAPITranscription(unittest.TestCase):
             # Should have been called with "en" (converted from "en-us")
             mock_server.assert_called_once_with(ANY, "en", ANY, ANY)
 
+    def test_transcribe_with_language_en_in_conversion(self):
+        """English (India) catalog id must map to Whisper code 'en'."""
+        from unittest.mock import patch
+
+        with patch.object(self.manager, "_try_whispercpp_server_api") as mock_server:
+            mock_server.return_value = "test"
+
+            self.manager.language = "en-in"
+            self.manager._transcribe_with_remote_api([b"audio"], self.manager._http_session)
+
+            mock_server.assert_called_once_with(ANY, "en", ANY, ANY)
+
     def test_transcribe_with_language_auto(self):
         """Test with auto language detection."""
         from unittest.mock import patch

tests/test_speech_recognition.py

@@ -305,6 +305,35 @@ class TestSpeechRecognition(unittest.TestCase):
         self.assertIsNone(hu["vosk"])
         self.assertNotIn("hu", VOSK_MODEL_INFO["small"]["languages"])
 
+    def test_resolve_whisper_language_maps_english_variants(self):
+        """Catalog ids like en-us/en-in must become Whisper's 'en' code."""
+        from unittest.mock import patch
+
+        from vocalinux.speech_recognition.recognition_manager import resolve_whisper_language
+
+        self.assertEqual(resolve_whisper_language("en-us"), "en")
+        self.assertEqual(resolve_whisper_language("en-in"), "en")
+        self.assertEqual(resolve_whisper_language("hu"), "hu")
+        self.assertEqual(resolve_whisper_language("ja"), "ja")
+        self.assertIsNone(resolve_whisper_language("auto"))
+        # Unknown catalog ids pass through unchanged.
+        self.assertEqual(resolve_whisper_language("zz-unknown"), "zz-unknown")
+
+        # Legacy fallback when a code is missing from the catalog entirely.
+        with patch(
+            "vocalinux.speech_recognition.recognition_manager.SUPPORTED_LANGUAGES",
+            {},
+        ):
+            self.assertEqual(resolve_whisper_language("en-us"), "en")
+            self.assertEqual(resolve_whisper_language("fr"), "fr")
+
+        # Entry present but without a whisper field still falls through.
+        with patch(
+            "vocalinux.speech_recognition.recognition_manager.SUPPORTED_LANGUAGES",
+            {"legacy": {"name": "Legacy", "vosk": None}},
+        ):
+            self.assertEqual(resolve_whisper_language("legacy"), "legacy")
+
     def test_vosk_init_italian_medium_model_resolves_correctly(self):
         """Selecting Italian + the 'medium' VOSK model must not crash (issue #550)."""
         manager = SpeechRecognitionManager(engine="vosk", language="it", model_size="medium")