Skip to content

Remove logic for overwriting speaker IDs (#72) #73

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Feb 22, 2023
Merged

Remove logic for overwriting speaker IDs (#72) #73

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
29 changes: 2 additions & 27 deletions detection/component_util/mpf_component_util/job_config.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -131,7 +131,6 @@ class DynamicSpeechJobConfig(TriggeredJobConfig):
:ivar stop_time: Stop time of the audio (in milliseconds)
:ivar fps: Frames per second for video jobs
:ivar speaker_id_prefix: Prefix for LONG_SPEAKER_ID
:ivar overwrite_ids: Whether this has been identified as a likely subjob
:ivar override_default_language: A default ISO 639-3 language to use when
languages defined in the speaker are not supported. If the feed-forward
track does not exist, this is None
Expand All @@ -143,7 +142,6 @@ def __init__(self, job: MpfSpeechJob):
self.stop_time: Optional[int] = None
self.fps: Optional[float] = None
self.speaker_id_prefix: str
self.overwrite_ids: bool
super().__init__(job)

# Properties related to dynamic speech pipelines
Expand All @@ -154,7 +152,6 @@ def __init__(self, job: MpfSpeechJob):

def _add_job_data(self, job: MpfSpeechJob):
super()._add_job_data(job)
self.overwrite_ids = self.is_triggered_job

media_duration = float(job.media_properties.get('DURATION', -1))
if isinstance(job, mpf.VideoJob):
Expand All @@ -173,14 +170,6 @@ def _add_job_data(self, job: MpfSpeechJob):
fpms = self.fps / 1000.0
media_frame_count = int(job.media_properties.get('FRAME_COUNT', -1))

# Determine whether this may be a subjob, based on whether the
# start and stop frames line up with the media length
if stop_frame is not None:
if (start_frame > 0
or (media_duration > 0 and (stop_frame / fpms) < media_duration)
or (media_frame_count > 0 and stop_frame < media_frame_count - 1)):
self.overwrite_ids = True

# Convert frame locations to timestamps
self.start_time = int(start_frame / fpms)

Expand All @@ -203,28 +192,14 @@ def _add_job_data(self, job: MpfSpeechJob):
self.stop_time = None
self.speaker_id_prefix = f"{self.start_time}-{self.stop_time or 'EOF'}-"

# Determine whether this may be a subjob, based on whether the
# start and stop frames line up with the media length
if self.stop_time is not None:
if self.start_time > 0 or (media_duration > 0 and self.stop_time < media_duration):
self.overwrite_ids = True

def _add_feed_forward_properties(self, job: MpfSpeechJob):
feed_forward_properties = job.feed_forward_track.detection_properties
speaker_id = mpf_util.get_property(
properties=feed_forward_properties,
key='SPEAKER_ID',
default_value='0',
key='LONG_SPEAKER_ID',
default_value='0-EOF-1',
prop_type=str
)
# If speaker ID was overwritten, use long speaker ID
if speaker_id == '0':
speaker_id = mpf_util.get_property(
properties=feed_forward_properties,
key='LONG_SPEAKER_ID',
default_value='0',
prop_type=str
)

gender = mpf_util.get_property(
properties=feed_forward_properties,
Expand Down