Skip to content

Cache + filter: filter is reapplied to cached file and saved #282

@christerdk

Description

@christerdk

System Health details

`## System Information

version core-2025.7.1
installation_type Home Assistant OS
dev false
hassio true
docker true
container_arch amd64
user root
virtualenv false
python_version 3.13.3
os_name Linux
os_version 6.12.23-haos
arch x86_64
timezone Asia/Bangkok
config_dir /config
Home Assistant Community Store
GitHub API ok
GitHub Content ok
GitHub Web ok
HACS Data ok
GitHub API Calls Remaining 5000
Installed Version 2.0.5
Stage running
Available Repositories 2112
Downloaded Repositories 21
Home Assistant Cloud
logged_in false
can_reach_cert_server ok
can_reach_cloud_auth ok
can_reach_cloud ok
Home Assistant Supervisor
host_os Home Assistant OS 15.2
update_channel stable
supervisor_version supervisor-2025.07.1
agent_version 1.7.2
docker_version 28.0.4
disk_total 30.8 GB
disk_used 14.1 GB
healthy true
supported true
host_connectivity true
supervisor_connectivity true
ntp_synchronized true
virtualization microsoft
board ova
supervisor_api ok
version_api ok
installed_addons File editor (5.8.0), Node-RED (20.0.0), Cloudflared (5.2.11), Zigbee2MQTT (2.5.1-1), ESPHome Device Builder (2025.6.3), Studio Code Server (5.19.3), Mosquitto broker (6.5.0), Samba share (12.5.2), Get HACS (1.3.1), Music Assistant Server (2.5.5), Piper (1.6.2)
Dashboards
dashboards 6
resources 8
views 13
mode storage
Network Configuration
adapters lo (disabled), eth0 (enabled, default, auto), docker0 (disabled), hassio (disabled), veth1749a84 (disabled), veth4451fbf (disabled), veth0edaa55 (disabled), vethbcf00e1 (disabled), veth63ddcab (disabled), veth1940113 (disabled), vethdede1fe (disabled), veth0215d43 (disabled), vethe2c6fce (disabled), veth3071b9b (disabled), vetha329d95 (disabled)
ipv4_addresses lo (127.0.0.1/8), eth0 (192.168.1.55/24), docker0 (172.30.232.1/23), hassio (172.30.32.1/23), veth1749a84 (), veth4451fbf (), veth0edaa55 (), vethbcf00e1 (), veth63ddcab (), veth1940113 (), vethdede1fe (), veth0215d43 (), vethe2c6fce (), veth3071b9b (), vetha329d95 ()
ipv6_addresses lo (::1/128), eth0 (2001:fb1:42:72c9:254d:b902:85f8:ea20/64, fe80::7615:4204:2ec3:2a5/64), docker0 (fe80::8c0a:7ff:fe1d:595d/64), hassio (fe80::6855:b7ff:fee5:b38d/64), veth1749a84 (fe80::144d:beff:fe7d:21d0/64), veth4451fbf (fe80::c801:3cff:fe78:1134/64), veth0edaa55 (fe80::c2d:cff:feca:1d55/64), vethbcf00e1 (fe80::bc7d:8aff:fe20:3ffd/64), veth63ddcab (fe80::3cd4:12ff:fe35:c981/64), veth1940113 (fe80::589f:8ff:fe9c:ef38/64), vethdede1fe (fe80::a82b:2eff:fe96:4e3e/64), veth0215d43 (fe80::c064:15ff:fe81:5bb5/64), vethe2c6fce (fe80::1413:36ff:fe89:eec3/64), veth3071b9b (fe80::d8c8:19ff:fe78:ef31/64), vetha329d95 (fe80::649f:a1ff:fe5e:80b7/64)
announce_addresses 192.168.1.55, 2001:fb1:42:72c9:254d:b902:85f8:ea20, fe80::7615:4204:2ec3:2a5
Recorder
oldest_recorder_run July 9, 2025 at 10:28 AM
current_recorder_run July 21, 2025 at 8:37 PM
estimated_db_size 74.07 MiB
database_engine sqlite
database_version 3.48.0
`

Checklist

  • I have enabled debug logging for my installation.
  • I have filled out the issue template to the best of my ability.
  • This issue only contains 1 issue (if you have multiple issues, open one issue for each issue).
  • This issue is not a duplicate issue of currently previous issues..

Describe the issue

I have set a audio conversion like this: -filter:a volume=1.6

Sounds great first time around. But I noticed something when I have Chime TTS repeat the same message several times: the message just gets louder and louder.

Chime TTS Version v1.2.2

This seems to be related: #128

Reproduction steps

  1. Set up a chime TTS action (chime_tts.say)
  2. Set cache = true
  3. Set audio -filter:a volume=1.6
  4. Run the Chime action X times in the same script. I run the same message 4 times ("We have visitors!") as part of a doorbell announcement automation.

Observation: it seems that the cached version includes the audio conversion (which can be good) but also that the audio conversion is added again to the cached version (which can be considered bad).

If this is the case, it seems that there are two different ways to approach this: (1) save the audio before the audio is processed and apply at every playback or (2) save the process audio but make sure it doesn't get applied again on the cached file. I think the latter is what most would expect.

Debug logs

.

Diagnostics dump

No response

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions