Start adding config for audio

This commit is contained in:
Nick Mowen 2023-06-17 16:20:42 -06:00
parent 8055fbc6e8
commit e43a7e65f9
3 changed files with 176 additions and 97 deletions

View File

@ -1,101 +1,101 @@
Speech
Child speech, kid speaking
Conversation
Narration, monologue
Babbling
Speech synthesizer
Shout
Bellow
Whoop
Yell
Children shouting
Screaming
speech
speech
speech
speech
babbling
speech
yell
bellow
whoop
yell
yell
yell
Whispering
Laughter
Baby laughter
Giggle
Snicker
Belly laugh
Chuckle, chortle
Crying, sobbing
Baby cry, infant cry
Whimper
Wail, moan
Sigh
Singing
Choir
Yodeling
Chant
Mantra
Child singing
Synthetic singing
Rapping
Humming
Groan
Grunt
Whistling
Breathing
Wheeze
Snoring
Gasp
Pant
Snort
Cough
Throat clearing
Sneeze
Sniff
Run
Shuffle
Walk, footsteps
Chewing, mastication
Biting
Gargling
Stomach rumble
Burping, eructation
Hiccup
Fart
Hands
Finger snapping
Clapping
Heart sounds, heartbeat
Heart murmur
Cheering
Applause
Chatter
Crowd
Hubbub, speech noise, speech babble
Children playing
Animal
Domestic animals, pets
Dog
Bark
Yip
Howl
Bow-wow
Growling
Whimper (dog)
Cat
Purr
Meow
Hiss
Caterwaul
Livestock, farm animals, working animals
Horse
Clip-clop
Neigh, whinny
Cattle, bovinae
Moo
Cowbell
Pig
Oink
Goat
Bleat
Sheep
Fowl
Chicken, rooster
Cluck
Crowing, cock-a-doodle-doo
Turkey
laughter
laughter
laughter
snicker
laughter
laughter
crying
crying
crying
yell
sigh
singing
choir
sodeling
chant
mantra
child_singing
synthetic_singing
rapping
humming
groan
grunt
whistling
breathing
wheeze
snoring
gasp
pant
snort
cough
throat clearing
sneeze
sniff
run
shuffle
footsteps
chewing
biting
gargling
stomach_rumble
burping
hiccup
fart
hands
finger snapping
clapping
heartbeat
heart_murmur
cheering
applause
chatter
crowd
speech
children_playing
animal
pets
dog
bark
yip
howl
bow-wow
growling
whimper_dog
cat
purr
meow
hiss
caterwaul
livestock
horse
clip-clop
neigh
cattle
moo
cowbell
pig
oink
goat
bleat
sheep
fowl
chicken
cluck
cock-a-doodle-doo
turkey
Gobble
Duck
Quack

View File

@ -40,6 +40,7 @@ DEFAULT_TIME_FORMAT = "%m/%d/%Y %H:%M:%S"
FRIGATE_ENV_VARS = {k: v for k, v in os.environ.items() if k.startswith("FRIGATE_")}
DEFAULT_TRACKED_OBJECTS = ["person"]
DEFAULT_LISTEN_AUDIO = ["bark", "speech", "yell", "scream"]
DEFAULT_DETECTORS = {"cpu": {"type": "cpu"}}
@ -387,6 +388,16 @@ class ObjectConfig(FrigateBaseModel):
mask: Union[str, List[str]] = Field(default="", title="Object mask.")
class AudioConfig(FrigateBaseModel):
enabled: bool = Field(default=False, title="Enable audio events.")
max_not_heard: int = Field(
default=30, title="Seconds of not hearing the type of audio to end the event."
)
listen: List[str] = Field(
default=DEFAULT_LISTEN_AUDIO, title="Audio to listen for."
)
class BirdseyeModeEnum(str, Enum):
objects = "objects"
motion = "motion"
@ -466,6 +477,7 @@ class FfmpegConfig(FrigateBaseModel):
class CameraRoleEnum(str, Enum):
audio = "audio"
record = "record"
rtmp = "rtmp"
detect = "detect"
@ -627,6 +639,9 @@ class CameraConfig(FrigateBaseModel):
objects: ObjectConfig = Field(
default_factory=ObjectConfig, title="Object configuration."
)
audio: AudioConfig = Field(
default_factory=AudioConfig, title="Audio events configuration."
)
motion: Optional[MotionConfig] = Field(title="Motion detection configuration.")
detect: DetectConfig = Field(
default_factory=DetectConfig, title="Object detection configuration."
@ -657,12 +672,16 @@ class CameraConfig(FrigateBaseModel):
# add roles to the input if there is only one
if len(config["ffmpeg"]["inputs"]) == 1:
has_rtmp = "rtmp" in config["ffmpeg"]["inputs"][0].get("roles", [])
has_audio = "audio" in config["audio"]["inputs"][0].get("roles", [])
config["ffmpeg"]["inputs"][0]["roles"] = [
"record",
"detect",
]
if has_audio:
config["ffmpeg"]["inputs"][0].append("audio")
if has_rtmp:
config["ffmpeg"]["inputs"][0]["roles"].append("rtmp")
@ -795,6 +814,11 @@ def verify_config_roles(camera_config: CameraConfig) -> None:
f"Camera {camera_config.name} has rtmp enabled, but rtmp is not assigned to an input."
)
if camera_config.audio.enabled and "audio" not in assigned_roles:
raise ValueError(
f"Camera {camera_config.name} has audio events enabled, but audio is not assigned to an input."
)
def verify_valid_live_stream_name(
frigate_config: FrigateConfig, camera_config: CameraConfig
@ -907,6 +931,9 @@ class FrigateConfig(FrigateBaseModel):
objects: ObjectConfig = Field(
default_factory=ObjectConfig, title="Global object configuration."
)
audio: AudioConfig = Field(
default_factory=AudioConfig, title="Global Audio events configuration."
)
motion: Optional[MotionConfig] = Field(
title="Global motion detection configuration."
)
@ -931,6 +958,7 @@ class FrigateConfig(FrigateBaseModel):
# Global config to propagate down to camera level
global_config = config.dict(
include={
"audio": ...,
"birdseye": ...,
"record": ...,
"snapshots": ...,

51
frigate/events/audio.py Normal file
View File

@ -0,0 +1,51 @@
"""Handle creating audio events."""
import logging
import multiprocessing as mp
import signal
import threading
from types import FrameType
from typing import Optional
from setproctitle import setproctitle
from frigate.config import AudioConfig, CameraConfig, FrigateConfig
from frigate.const import CACHE_DIR
from frigate.util import listen
logger = logging.getLogger(__name__)
FFMPEG_COMMAND = "ffmpeg -vn -i {} -f s16le -ar 16000 -ac 1 -y {}"
def listen_to_audio(config: FrigateConfig, event_queue: mp.Queue) -> None:
stop_event = mp.Event()
def receiveSignal(signalNumber: int, frame: Optional[FrameType]) -> None:
stop_event.set()
signal.signal(signal.SIGTERM, receiveSignal)
signal.signal(signal.SIGINT, receiveSignal)
threading.current_thread().name = "process:recording_manager"
setproctitle("frigate.recording_manager")
listen()
for camera in config.cameras.values():
if camera.enabled and camera.audio.enabled:
AudioEventMaintainer(camera, stop_event)
class AudioEventMaintainer(threading.Thread):
def __init__(self, camera: CameraConfig, stop_event: mp.Event) -> None:
threading.Thread.__init__(self)
self.name = f"{camera.name}_audio_event_processor"
self.config = camera
self.stop_event = stop_event
def run(self) -> None:
self.pipe = f"{CACHE_DIR}/{self.config.name}-audio"
self.ffmpeg_command = FFMPEG_COMMAND.format(
[i.path for i in self.config.ffmpeg.inputs if "audio" in i.roles][0],
self.pipe,
)