jpandeinge commited on
Commit
9c4c263
·
1 Parent(s): a889160

another repo test

Browse files
Files changed (1) hide show
  1. app.py +5 -5
app.py CHANGED
@@ -19,7 +19,7 @@ GREETINGS_DIR = './greetings'
19
  greeting_files = [f.name for f in os.scandir(GREETINGS_DIR)]
20
 
21
 
22
- DATASET_REPO_URL = "https://huggingface.co/datasets/meyabase/crowdsource-oshiwambo-speech-greetings"
23
  REPOSITORY_DIR = "data"
24
  LOCAL_DIR = 'data_local'
25
  os.makedirs(LOCAL_DIR,exist_ok=True)
@@ -89,7 +89,7 @@ def save_record(language,record,greeting,gender,accent,greeting_history,current_
89
 
90
  _ = upload_file(path_or_fileobj = audio_output_filename,
91
  path_in_repo =repo_audio_path,
92
- repo_id='meyabase/crowdsource-oshiwambo-speech-greetings',
93
  repo_type='dataset',
94
  token=HF_TOKEN
95
  )
@@ -98,7 +98,7 @@ def save_record(language,record,greeting,gender,accent,greeting_history,current_
98
  repo_json_path = os.path.join(REPOSITORY_DIR,os.path.join(audio_name,'metadata.jsonl'))
99
  _ = upload_file(path_or_fileobj = json_file_path,
100
  path_in_repo =repo_json_path,
101
- repo_id='meyabase/crowdsource-oshiwambo-speech-greetings',
102
  repo_type='dataset',
103
  token=HF_TOKEN
104
  )
@@ -166,7 +166,7 @@ def get_metadata_of_dataset():
166
  repo_recordings = [os.path.join(REPOSITORY_DATA_DIR,f.name) for f in os.scandir(REPOSITORY_DATA_DIR)] if os.path.isdir(REPOSITORY_DATA_DIR) else []
167
 
168
  audio_repo = [os.path.join(f,'audio.wav') for f in repo_recordings]
169
- audio_repo = [a.replace('data/data/','https://huggingface.co/datasets/meyabase/crowdsource-oshiwambo-speech-greetings/resolve/main/data/') for a in audio_repo]
170
  metadata_all = [get_metadata_json(os.path.join(f,'metadata.jsonl')) for f in repo_recordings]
171
  metadata_all = [m for m in metadata_all if m!=[]]
172
  return metadata_all
@@ -177,7 +177,7 @@ def display_records():
177
  repo_recordings = [os.path.join(REPOSITORY_DATA_DIR,f.name) for f in os.scandir(REPOSITORY_DATA_DIR)] if os.path.isdir(REPOSITORY_DATA_DIR) else []
178
 
179
  audio_repo = [os.path.join(f,'audio.wav') for f in repo_recordings]
180
- audio_repo = [a.replace('data/data/','https://huggingface.co/datasets/meyabase/crowdsource-oshiwambo-speech-greetings/resolve/main/data/') for a in audio_repo]
181
  metadata_repo = [read_json_lines(os.path.join(f,'metadata.jsonl'))[0] for f in repo_recordings]
182
  audios_all = audio_repo
183
  metadata_all = metadata_repo
 
19
  greeting_files = [f.name for f in os.scandir(GREETINGS_DIR)]
20
 
21
 
22
+ DATASET_REPO_URL = "https://huggingface.co/datasets/meyabase/crowd-oshiwambo-speech-greetings"
23
  REPOSITORY_DIR = "data"
24
  LOCAL_DIR = 'data_local'
25
  os.makedirs(LOCAL_DIR,exist_ok=True)
 
89
 
90
  _ = upload_file(path_or_fileobj = audio_output_filename,
91
  path_in_repo =repo_audio_path,
92
+ repo_id='meyabase/crowd-oshiwambo-speech-greetings',
93
  repo_type='dataset',
94
  token=HF_TOKEN
95
  )
 
98
  repo_json_path = os.path.join(REPOSITORY_DIR,os.path.join(audio_name,'metadata.jsonl'))
99
  _ = upload_file(path_or_fileobj = json_file_path,
100
  path_in_repo =repo_json_path,
101
+ repo_id='meyabase/crowd-oshiwambo-speech-greetings',
102
  repo_type='dataset',
103
  token=HF_TOKEN
104
  )
 
166
  repo_recordings = [os.path.join(REPOSITORY_DATA_DIR,f.name) for f in os.scandir(REPOSITORY_DATA_DIR)] if os.path.isdir(REPOSITORY_DATA_DIR) else []
167
 
168
  audio_repo = [os.path.join(f,'audio.wav') for f in repo_recordings]
169
+ audio_repo = [a.replace('data/data/','https://huggingface.co/datasets/meyabase/crowd-oshiwambo-speech-greetings/resolve/main/data/') for a in audio_repo]
170
  metadata_all = [get_metadata_json(os.path.join(f,'metadata.jsonl')) for f in repo_recordings]
171
  metadata_all = [m for m in metadata_all if m!=[]]
172
  return metadata_all
 
177
  repo_recordings = [os.path.join(REPOSITORY_DATA_DIR,f.name) for f in os.scandir(REPOSITORY_DATA_DIR)] if os.path.isdir(REPOSITORY_DATA_DIR) else []
178
 
179
  audio_repo = [os.path.join(f,'audio.wav') for f in repo_recordings]
180
+ audio_repo = [a.replace('data/data/','https://huggingface.co/datasets/meyabase/crowd-oshiwambo-speech-greetings/resolve/main/data/') for a in audio_repo]
181
  metadata_repo = [read_json_lines(os.path.join(f,'metadata.jsonl'))[0] for f in repo_recordings]
182
  audios_all = audio_repo
183
  metadata_all = metadata_repo