zev/tatt
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fixed #49, fixed #52

Browse Source
This commit is contained in:
zevav
2019-03-08 11:35:05 -05:00
parent 9952f5fe4b
commit 82efb77ff2
3 changed files with 6311 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6293
GO_EMC_100914_Interview_first_minute_phone_call_model.mp3.txt Normal file
View File

File diff suppressed because it is too large Load Diff

19
tatt/transcribe.py
View File

@@ -90,15 +90,22 @@ def status(job_name):
@cli.command()
@click.option('--punctuation', is_flag=True, default=True,
help='only for Google Speech, defaults to True')
@click.option('--speaker-id', is_flag=True, default=True,
help='only for Google Speech, defaults to True')
@click.option('--speaker-id/--no-speaker-id', is_flag=True, default=True,
help='only for google and amazon, defaults to True')
@click.option('--num_speakers', default=2, type=int,
help='only for google and amazon, defaults to 2')
@click.option('--model', default='phone_call',
help='only for Google Speech, defaults to "phone_call"')
@click.option('--use-enhanced', is_flag=True, default=True,
help='only for Google Speech, defaults to True')
@click.argument('media_filepath', type=str)
@click.argument('service_name', type=str)
def this(media_filepath, service_name, punctuation, speaker_id, model,
def this(media_filepath,
service_name,
punctuation,
speaker_id,
num_speakers,
model,
use_enhanced):
"""Sends a media file to be transcribed."""
if service_name == 'google':
@@ -107,6 +114,12 @@ def this(media_filepath, service_name, punctuation, speaker_id, model,
enable_speaker_diarization=speaker_id,
model=model,
use_enhanced=use_enhanced,
num_speakers=num_speakers,
)
elif service_name == 'amazon':
transcribe_kwargs = dict(
enable_speaker_diarization=speaker_id,
num_speakers=num_speakers,
)
else:
transcribe_kwargs = {}

2
tatt/vendors/google.py vendored
View File

@@ -100,6 +100,7 @@ class Transcriber(TranscriberBaseClass):
language_code='en-US',
enable_automatic_punctuation=True,
enable_speaker_diarization=True,
num_speakers=2,
model='phone_call',
use_enhanced=True,
) -> str:
@@ -124,6 +125,7 @@ class Transcriber(TranscriberBaseClass):
language_code=language_code,
enable_automatic_punctuation=enable_automatic_punctuation,
enable_speaker_diarization=enable_speaker_diarization,
diarization_speaker_count=num_speakers,
model=model,
use_enhanced=use_enhanced,
)