more efficient birndet_analysis.service
more efficient extraction.service cleaner birdnet_analysis logging cleaner birdnet.conf functioning species_notifier
This commit is contained in:
+29
-34
@@ -1,17 +1,14 @@
|
||||
#!/usr/bin/env bash
|
||||
# Runs BirdNET in virtual environment
|
||||
# Runs BirdNET-Lite
|
||||
#set -x
|
||||
source /etc/birdnet/birdnet.conf
|
||||
CUSTOM_LIST="/home/pi/BirdNET-Lite/custom_species_list.txt"
|
||||
DAYS=(
|
||||
"today"
|
||||
)
|
||||
|
||||
# Create an array of the day's audio files
|
||||
# Uses 1st argument:
|
||||
# Create an array of the audio files
|
||||
# Takes one argument:
|
||||
# - {DIRECTORY}
|
||||
get_files() {
|
||||
echo "Starting get_files() for ${1}"
|
||||
echo "get_files() for ${1:19}"
|
||||
files=($( find ${1} -maxdepth 1 -name '*wav' \
|
||||
| sort \
|
||||
| awk -F "/" '{print $NF}' ))
|
||||
@@ -20,33 +17,33 @@ get_files() {
|
||||
|
||||
# Move all files that have been analyzed already into newly created "Analyzed"
|
||||
# directory
|
||||
# Uses 1st argument:
|
||||
# Takes one argument:
|
||||
# - {DIRECTORY}
|
||||
move_analyzed() {
|
||||
echo "Starting move_analyzed() for ${1}"
|
||||
echo "Starting move_analyzed() for ${1:19}"
|
||||
for i in "${files[@]}";do
|
||||
j="$(echo "${i}" | cut -d'.' -f1-2).csv"
|
||||
if [ -f "${1}/${j}" ];then
|
||||
if [ ! -d "${1}-Analyzed" ];then
|
||||
mkdir -vvvvvvvp "${1}-Analyzed" && echo "'Analyzed' directory created"
|
||||
j="${i}.csv"
|
||||
if [ -f "${1}/${j}" ];then
|
||||
if [ ! -d "${1}-Analyzed" ];then
|
||||
mkdir -p "${1}-Analyzed" && echo "'Analyzed' directory created"
|
||||
fi
|
||||
echo "Moving analyzed files to new directory"
|
||||
mv "${1}/${i}" "${1}-Analyzed/"
|
||||
mv "${1}/${j}" "${1}-Analyzed/"
|
||||
fi
|
||||
echo "Moving analyzed files to new directory"
|
||||
mv "${1}/${i}" "${1}-Analyzed/"
|
||||
mv "${1}/${j}" "${1}-Analyzed/"
|
||||
fi
|
||||
done
|
||||
done
|
||||
}
|
||||
|
||||
# Run BirdNET analysis on the remaining WAVE files for the day
|
||||
# Uses 1st and 2nd arguments:
|
||||
# Run BirdNET-Lite on the WAVE files from get_files()
|
||||
# Uses one argument:
|
||||
# - {DIRECTORY}
|
||||
# - {"today", "yesterday", "2 days ago",...}
|
||||
run_analysis() {
|
||||
echo "Starting run_analysis() for ${1}"
|
||||
echo "Starting run_analysis() for ${1:19}"
|
||||
WEEK=$(date --date="${2}" +"%U")
|
||||
cd ${HOME}/BirdNET-Lite || exit 1
|
||||
for i in "${files[@]}";do
|
||||
if [ -f ${1}/${i} ] && [ ! -f ${CUSTOM_LIST} ];then
|
||||
set -x
|
||||
python3 analyze.py \
|
||||
--i "${1}/${i}" \
|
||||
--o "${1}/${i}.csv" \
|
||||
@@ -55,7 +52,9 @@ run_analysis() {
|
||||
--week "${WEEK}" \
|
||||
--overlap "${OVERLAP}" \
|
||||
--min_conf "${CONFIDENCE}"
|
||||
set +x
|
||||
elif [ -f ${1}/${i} ] && [ -f ${CUSTOM_LIST} ];then
|
||||
set -x
|
||||
python3 analyze.py \
|
||||
--i "${1}/${i}" \
|
||||
--o "${1}/${i}.csv" \
|
||||
@@ -65,6 +64,7 @@ run_analysis() {
|
||||
--overlap "${OVERLAP}" \
|
||||
--min_conf "${CONFIDENCE}" \
|
||||
--custom_list "${CUSTOM_LIST}"
|
||||
set +x
|
||||
fi
|
||||
done
|
||||
}
|
||||
@@ -74,22 +74,17 @@ run_analysis() {
|
||||
# - {DIRECTORY}
|
||||
# - {"today", "yesterday", "2 days ago",...}
|
||||
run_birdnet() {
|
||||
echo "Starting run_birdnet() in \"${1}\" for \""${2}"\""
|
||||
sleep 1
|
||||
echo "Starting run_birdnet() for \"${1:19}\""
|
||||
get_files "${1}"
|
||||
sleep 1
|
||||
move_analyzed "${1}"
|
||||
sleep 1
|
||||
run_analysis "${1}" "${2}"
|
||||
run_analysis "${1}"
|
||||
}
|
||||
|
||||
if [ $(find ${RECS_DIR} -maxdepth 1 -name '*wav' | wc -l) -gt 0 ];then
|
||||
run_birdnet "${RECS_DIR}" "today"
|
||||
run_birdnet "${RECS_DIR}"
|
||||
fi
|
||||
|
||||
for i in ${!DAYS[@]};do
|
||||
DIRECTORY="$RECS_DIR/$(date --date="${DAYS[$i]}" "+%B-%Y/%d-%A")"
|
||||
if [ $(find ${DIRECTORY} -name '*wav' | wc -l) -gt 0 ];then
|
||||
run_birdnet "${DIRECTORY}" "${DAYS[$i]}"
|
||||
fi
|
||||
done
|
||||
DIRECTORY="$RECS_DIR/$(date "+%B-%Y/%d-%A")"
|
||||
if [ $(find ${DIRECTORY} -name '*wav' | wc -l) -gt 0 ];then
|
||||
run_birdnet "${DIRECTORY}"
|
||||
fi
|
||||
|
||||
@@ -12,10 +12,10 @@ if pgrep arecord &> /dev/null ;then
|
||||
echo "Recording"
|
||||
else
|
||||
if [ -z ${REC_CARD} ];then
|
||||
arecord -f S16_LE -c${CHANNELS} -r48000 -t wav --max-file-time 9\
|
||||
arecord -f S16_LE -c${CHANNELS} -r48000 -t wav --max-file-time 6\
|
||||
--use-strftime ${RECS_DIR}/%B-%Y/%d-%A/%F-birdnet-${STAMP}.wav
|
||||
else
|
||||
arecord -f S16_LE -c${CHANNELS} -r48000 -t wav --max-file-time 9\
|
||||
arecord -f S16_LE -c${CHANNELS} -r48000 -t wav --max-file-time 6\
|
||||
-D "${REC_CARD}" --use-strftime \
|
||||
${RECS_DIR}/%B-%Y/%d-%A/%F-birdnet-${STAMP}.wav
|
||||
fi
|
||||
|
||||
@@ -13,6 +13,7 @@ source /etc/birdnet/birdnet.conf
|
||||
|
||||
# Set Variables
|
||||
TMPFILE=$(mktemp)
|
||||
ANALYZED=${RECS_DIR}/*/*Analyzed
|
||||
# SCAN_DIRS are all directories marked "Analyzed"
|
||||
SCAN_DIRS=($(find ${ANALYZED} -type d | sort ))
|
||||
|
||||
|
||||
@@ -411,15 +411,7 @@ CONFIDENCE="0.7"
|
||||
|
||||
CHANNELS=${CHANNELS}
|
||||
|
||||
# Don't touch the three below
|
||||
|
||||
## ANALYZED is where the extraction.service looks for audio and
|
||||
## BirdNET.selection.txt files after they have been processed by the
|
||||
## birdnet_analysis.service. This is NOT where the analyzed files are moved --
|
||||
## analyzed files are always created within the same directory
|
||||
## birdnet_analysis.service finds them.
|
||||
|
||||
ANALYZED=${RECS_DIR}/*/*Analyzed
|
||||
# Don't touch the variables below
|
||||
|
||||
## SYSTEMD_MOUNT is created from the RECS_DIR variable to comply with systemd
|
||||
## mount naming requirements.
|
||||
|
||||
@@ -23,7 +23,7 @@ Description=BirdNET Analysis
|
||||
Restart=always
|
||||
RuntimeMaxSec=10800
|
||||
Type=simple
|
||||
RestartSec=1
|
||||
RestartSec=2
|
||||
User=${USER}
|
||||
ExecStart=/usr/local/bin/birdnet_analysis.sh
|
||||
[Install]
|
||||
|
||||
@@ -10,7 +10,7 @@ TMPFILE=$(mktemp)
|
||||
[ -f ${IDFILE} ] || touch ${IDFILE}
|
||||
cat "${IDFILE}" > "${TMPFILE}"
|
||||
|
||||
/usr/local/bin/update_species.sh &> /dev/null
|
||||
/usr/local/bin/update_species.sh > /dev/null
|
||||
|
||||
if ! diff "${IDFILE}" "${TMPFILE}" &> /dev/null; then
|
||||
SPECIES=("$(diff "${IDFILE}" "${TMPFILE}" \
|
||||
@@ -19,9 +19,10 @@ if ! diff "${IDFILE}" "${TMPFILE}" &> /dev/null; then
|
||||
| awk '{for(i=4;i<=NF;++i)printf $i""FS ; print ""}')")
|
||||
|
||||
NOTIFICATION="New Species Detection: "${SPECIES[@]}""
|
||||
|
||||
echo "Sending the following notification:
|
||||
${NOTIFICATION}"
|
||||
if [ ! -z ${PUSHED_APP_KEY} ];then
|
||||
curl -X POST -s \
|
||||
curl -X POST \
|
||||
--form-string "app_key=${PUSHED_APP_KEY}" \
|
||||
--form-string "app_secret=${PUSHED_APP_SECRET}" \
|
||||
--form-string "target_type=app" \
|
||||
|
||||
@@ -9,8 +9,6 @@ TMPFILE=$(mktemp) || exit 1
|
||||
|
||||
[ -f ${IDFILE} ] || touch ${IDFILE}
|
||||
|
||||
IDFILEBAKUP="${IDFILE}.bak"
|
||||
|
||||
if [ $(find ${PROCESSED} -name '*csv' | wc -l) -ge 1 ];then
|
||||
sort $(find ${PROCESSED} ${ANALYZED} ${EXTRACTED} -name '*csv') \
|
||||
| awk -F\; '!/Scientific/ {print"Common Name: " $4 "\nScientific Name: " $3""}' \
|
||||
|
||||
Reference in New Issue
Block a user