towhee
/
audio-embedding-vggish
copied
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Readme
Files and versions
106 lines
2.3 KiB
106 lines
2.3 KiB
name: 'audio_decode'
|
|
operators:
|
|
-
|
|
name: '_start_op'
|
|
function: '_start_op'
|
|
init_args:
|
|
inputs:
|
|
-
|
|
df: '_start_df'
|
|
name: 'audio_path'
|
|
col: 0
|
|
outputs:
|
|
-
|
|
df: 'input_df'
|
|
iter_info:
|
|
type: map
|
|
-
|
|
name: 'audio_decoder'
|
|
function: 'towhee/audio-decoder'
|
|
init_args:
|
|
inputs:
|
|
-
|
|
df: 'input_df'
|
|
name: 'audio_path'
|
|
col: 0
|
|
outputs:
|
|
-
|
|
df: 'audio_df'
|
|
iter_info:
|
|
type: generator
|
|
-
|
|
name: 'embedding_model'
|
|
function: 'towhee/torch-vggish'
|
|
init_args:
|
|
inputs:
|
|
-
|
|
df: 'audio_df'
|
|
name: 'audio'
|
|
col: 0
|
|
-
|
|
df: 'audio_df'
|
|
name: 'sample_rate'
|
|
col: 1
|
|
-
|
|
df: 'audio_df'
|
|
name: 'layout'
|
|
col: 2
|
|
outputs:
|
|
-
|
|
df: 'output_df'
|
|
iter_info:
|
|
type: time_window
|
|
params:
|
|
time_range_sec: 600
|
|
time_step_sec: 600
|
|
-
|
|
name: '_end_op'
|
|
function: '_end_op'
|
|
init_args:
|
|
inputs:
|
|
-
|
|
df: 'output_df'
|
|
name: 'vec'
|
|
col: 0
|
|
outputs:
|
|
-
|
|
df: '_end_df'
|
|
iter_info:
|
|
type: map
|
|
dataframes:
|
|
-
|
|
name: '_start_df'
|
|
columns:
|
|
-
|
|
name: 'audio_path'
|
|
vtype: 'str'
|
|
-
|
|
name: 'input_df'
|
|
columns:
|
|
-
|
|
name: 'audio_path'
|
|
vtype: 'str'
|
|
-
|
|
name: 'audio_df'
|
|
columns:
|
|
-
|
|
name: 'audio_frame'
|
|
vtype: 'numpy.ndarray'
|
|
-
|
|
name: 'sample_rate'
|
|
vtype: 'numpy.ndarray'
|
|
-
|
|
name: 'layout'
|
|
vtype: 'str'
|
|
-
|
|
name: 'output_df'
|
|
columns:
|
|
-
|
|
name: 'vec'
|
|
vtype: 'numpy.ndarray'
|
|
-
|
|
name: '_end_df'
|
|
columns:
|
|
-
|
|
name: 'vec'
|
|
vtype: 'numpy.ndarray'
|