add package.stxt

gradio_asr_en_libri100.py

@@ -12,9 +12,9 @@
 import os
 from glob import glob
 from loguru import logger
-# import soundfile as sf
+import soundfile as sf
 # import librosa
-from scipy.io import wavfile
+# from scipy.io import wavfile
 import gradio as gr
 
 from espnet_model_zoo.downloader import ModelDownloader
@@ -53,9 +53,9 @@ logger.info('model loaded')
 
 def predict(wav_file):
     logger.info('wav file loaded')
-    # speech, rate = sf.read(wav_file)
+    speech, rate = sf.read(wav_file)
     # speech, rate = librosa.load(wav_file, sr=16000)
-    rate, speech = wavfile.read(wav_file)
+    # rate, speech = wavfile.read(wav_file)
     nbests = model(speech)
     text, *_ = nbests[0]
     logger.info('predicted')

packages.txt

@@ -0,0 +1,2 @@
+cmake
+libsndfile1

requirements.txt

@@ -2,5 +2,4 @@ espnet==0.10.6
 espnet_model_zoo==0.1.7
 gradio==2.7.5.2
 loguru==0.6.0
-librosa
-scipy==1.7.3
+soundfile