From 0600482fe5ffcd41d089ad1b48b5d3f3d65f01b0 Mon Sep 17 00:00:00 2001 From: Malar Kannan Date: Thu, 28 Dec 2017 13:37:27 +0530 Subject: [PATCH] generating segmentation for words --- speech_segmentgen.py | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/speech_segmentgen.py b/speech_segmentgen.py index 4b3010d..b03995a 100644 --- a/speech_segmentgen.py +++ b/speech_segmentgen.py @@ -57,7 +57,8 @@ class Delegate (NSObject): '''Called automatically when the application has launched''' print("App Launched!") # phrases = story_texts()#random.sample(story_texts(), 100) # - phrases = test_texts(30) + # phrases = test_texts(30) + phrases = story_words() # print(phrases) generate_audio(phrases) @@ -174,14 +175,19 @@ class SynthesizerQueue(object): def story_texts(): - # story_file = './inputs/all_stories_hs.json' story_file = './inputs/all_stories.json' stories_data = json.load(open(story_file)) - # text_list_dup = [t[0] for i in stories_data.values() for t in i] text_list_dup = [t for i in stories_data.values() for t in i] text_list = sorted(list(set(text_list_dup))) return text_list +def story_words(): + story_file = './inputs/all_stories_hs.json' + stories_data = json.load(open(story_file)) + text_list_dup = [t[0] for i in stories_data.values() for t in i] + text_list = sorted(list(set(text_list_dup))) + return text_list + def test_texts(count=10): word_list = [i.strip('\n_') for i in open('./inputs/wordlist.txt','r').readlines()] text_list = sorted(random.sample(list(set(word_list)),count))