Update phoneme.py

phoneme.py

@@ -5,17 +5,6 @@ from transformers import AutoProcessor, AutoModelForCTC, Wav2Vec2PhonemeCTCToken
 import librosa
 from itertools import groupby
 from datasets import load_dataset
-from phonemizer import phonemize
-from phonemizer.backend.espeak.wrapper import EspeakWrapper
-
-# PHONEMIZER_ESPEAK_LIBRARY="c:\Program Files\eSpeak NG\libespeak-ng.dll"
-# PHONEMIZER_ESPEAK_PATH="c:\Program Files\eSpeak NG"
-# ESPEAK_PATH = os.getenv("PHONEMIZER_ESPEAK_LIBRARY")
-# if ESPEAK_PATH is not None:
-#     EspeakWrapper.set_library(ESPEAK_PATH)
-# print(f"Loaded environment variables PHONEMIZER_ESPEAK_LIBRARY: {ESPEAK_PATH}")
-# print(f"Using espeak library: {EspeakWrapper.library_path}")
-
 
 # Load the model and processor
 # checkpoint = "bookbot/wav2vec2-ljspeech-gruut"
@@ -59,16 +48,6 @@ def text_to_phonemes(text: str) -> str:
     return phonemes
 
 
-def text_to_phonemes_2(text: str) -> str:
-    s_time = time.time()
-    """Convert text to phonemes using phonemizer."""
-    phonemes = phonemize(text, language="en-us", backend="espeak", strip=True)
-    # phonemes = tokenizer.phonemize(text)
-    e_time = time.time()
-    print(f"Execution time of text_to_phonemes_2: {e_time - s_time:.6f} seconds")
-    return phonemes
-
-
 def separate_characters(input_string):
     no_spaces = input_string.replace(" ", "")
     spaced_string = " ".join(no_spaces)
@@ -134,10 +113,7 @@ def test_sound():
     # Call the phoneme prediction function
     predicted_phonemes = predict_phonemes(audio_array)
     adjusted_phonemes = adjust_phonemes(predicted_phonemes)
-
-    # expected_phonemes_2 = text_to_phonemes_2(expected_transcript)
     print(f"Expected Phonemes: {expected_phonemes}")
-    # print(f"Expected Phonemes 2: {expected_phonemes_2}")
     print(f"Predicted Phonemes: {predicted_phonemes}")
     print(f"Adjusted Phonemes: {adjusted_phonemes}")