Help! My Voice Assistant hasn't been working since the last updates

Configuration Voice Assistant
1

Yesterday, I updated Home Assistant Core to 2026.4.0 (from some 2026.3.x version) and ESP Home Builder to 2026.3.2.
Since then, my Voice Assistant (ESP32-S3-N16R8 + Max98357 + Inmp441, Speech-to-Phrase, Microwakeword, Raspberry Pi 4, 8GB) hasn’t been working.
Is there any breaking change related to the Voice Assistant that I might have missed?

… Over the past few days, I’ve been working on getting notifications like “The doorbell rang”, “The phone rang,” etc., to play on this station, and I’ve also integrated a motion detector and an additional LED light strip, but nothing that should affect the voice assistant (although it’s possible I broke something software-wise). Everything worked fine until the update (but perhaps my changes didn’t take effect due to a lack of restarts beforehand?).

Previous behavior: Power on => Voice assistant light ring (12 LEDs) turns white, then goes out after a few seconds.

Say the wake word and command => Blue LEDs rotate first in one direction, then in the other, and then the response is heard through the speaker, and the LEDs turn off again.

Current behavior: Power on => Voice assistant light ring (12 LEDs) turns white, then after a few seconds all LEDs turn blue and remain blue (the light doesn’t turn off).

Saying the wakeword and command => The blue LEDs circle in one direction and don’t stop until the wakeword is said again. They don’t move in the other direction, and there’s no response!

The log files and activity logs don’t show anything unusual. How can I debug this? Where should I look?

The ESPHome log file shows a strange new entry (“Parent I2S bus not free”
and “Driver failed to start; retrying in 1 second”):

[00:37:36.206][D][sensor:124]: 'SZ BME680 Humidity' >> 47.9 %
[00:37:36.293][D][dallas.temp.sensor:054]: 'temp_dallas_schlafzimmer': Got Temperature=22.500000°C
[00:37:36.297][D][sensor:124]: 'temp_dallas_schlafzimmer' >> 22.5 °C
[00:37:38.190][D][binary_sensor:048]: 'Schlafzimmerstation Bewegungsmelder' >> ON
[00:37:39.106][D][micro_wake_word:325]: Detected 'Okay Nabu' with sliding average probability is 0.56 and max probability is 0.93
[00:37:39.107][D][media_player:113]: 'Schlafzimmer Media Player' - Setting
[00:37:39.107][D][media_player:117]:   Command: STOP
[00:37:39.107][D][media_player:126]:  Announcement: yes
[00:37:39.107][D][speaker_media_player:454]: State changed to ANNOUNCING
[00:37:39.111][D][speaker_media_player.pipeline:119]: Reading FLAC file type
[00:37:39.111][D][ring_buffer:034][ann_read]: Created ring buffer with size 100000
[00:37:39.178][D][sensor:124]: 'Sz BME680 IAQ' >> 72 IAQ
[00:37:39.182][D][sensor:124]: 'SZ BME680 CO2 Equivalent' >> 766.7 ppm
[00:37:39.196][D][speaker_media_player.pipeline:129]: Decoded audio has 1 channels, 48000 Hz sample rate, and 16 bits per sample
[00:37:39.200][D][sensor:124]: 'SZ BME680 Breath VOC Equivalent' >> 0.9 ppm
[00:37:39.200][D][sensor:124]: 'SZ BME680 Pressure' >> 1016.5 hPa
[00:37:39.204][D][sensor:124]: 'SZ BME680 Temperature' >> 20.22 °C
[00:37:39.207][D][sensor:124]: 'SZ BME680 Humidity' >> 47.8 %
[00:37:39.208][D][resampler_speaker:120]: Starting
[00:37:39.208][D][ring_buffer:034]: Created ring buffer with size 9600
[00:37:39.210][D][ring_buffer:034][resampler]: Created ring buffer with size 48000
[00:37:39.221][D][speaker_mixer:454]: Starting
[00:37:39.249][D][i2s_audio.speaker:102]: Starting
[00:37:39.250][E][i2s_audio.speaker:521]: Parent I2S bus not free
[00:37:39.251][E][i2s_audio.speaker:148]: Driver failed to start; retrying in 1 second
[00:37:39.254][E][component:420]: i2s_audio.speaker set Error flag: unspecified
[00:37:39.254][D][ring_buffer:034][speaker_task]: Created ring buffer with size 12800
[00:37:39.266][D][i2s_audio.speaker:106]: Started
[00:37:39.389][D][voice_assistant:483]: State changed from IDLE to START_MICROPHONE
[00:37:39.395][D][voice_assistant:490]: Desired state set to START_PIPELINE
[00:37:39.395][D][voice_assistant:208]: Starting Microphone
[00:37:39.395][D][ring_buffer:034]: Created ring buffer with size 16384
[00:37:39.396][D][voice_assistant:483]: State changed from START_MICROPHONE to STARTING_MICROPHONE
[00:37:39.398][D][voice_assistant:483]: State changed from STARTING_MICROPHONE to START_PIPELINE
[00:37:39.407][D][voice_assistant:229]: Requesting start
[00:37:39.409][D][voice_assistant:483]: State changed from START_PIPELINE to STARTING_PIPELINE
[00:37:39.420][D][voice_assistant:505]: Client started, streaming microphone
[00:37:39.422][D][voice_assistant:483]: State changed from STARTING_PIPELINE to STREAMING_MICROPHONE
[00:37:39.426][D][voice_assistant:490]: Desired state set to STREAMING_MICROPHONE
[00:37:39.441][D][voice_assistant:631]: Event Type: 1
[00:37:39.442][D][voice_assistant:634]: Assist Pipeline running
[00:37:39.446][D][voice_assistant:631]: Event Type: 3
[00:37:39.452][D][voice_assistant:653]: STT started
[00:37:39.453][D][light:079]: 'sz_voice_assistant_leds' Setting:
[00:37:39.456][D][light:092]:   State: ON
[00:37:39.466][D][light:067]:   Brightness: 20%
[00:37:39.471][D][light:153]:   Effect: 'Waiting for Command'
[00:37:39.501][D][speaker_media_player:454]: State changed to IDLE
[00:37:39.504][D][resampler_speaker:151]: Stopped
[00:37:40.068][D][binary_sensor:048]: 'Schlafzimmerstation Bewegungsmelder' >> OFF
[00:37:40.245][D][speaker_mixer:472]: Stopped
[00:37:40.255][E][component:433]: i2s_audio.speaker cleared Error flag
[00:37:40.484][D][i2s_audio.speaker:111]: Stopping
[00:37:40.487][D][i2s_audio.speaker:116]: Stopped
[00:37:40.488][D][voice_assistant:631]: Event Type: 11
[00:37:40.488][D][voice_assistant:834]: Starting STT by VAD
[00:37:40.495][D][light:079]: 'sz_voice_assistant_leds' Setting:
[00:37:40.498][D][light:067]:   Brightness: 20%
[00:37:40.499][D][light:153]:   Effect: 'Listening For Command'
[00:37:40.712][D][bh1750.sensor:157]: 'bh1750_illuminance_schlafzimmer': Illuminance=1.6lx
[00:37:40.713][D][sensor:124]: 'bh1750_illuminance_schlafzimmer' >> 1.585 lx
[00:37:42.095][D][voice_assistant:631]: Event Type: 12
[00:37:42.096][D][voice_assistant:838]: STT by VAD end
[00:37:42.096][D][voice_assistant:483]: State changed from STREAMING_MICROPHONE to STOP_MICROPHONE
[00:37:42.096][D][voice_assistant:490]: Desired state set to AWAITING_RESPONSE
[00:37:42.099][D][voice_assistant:483]: State changed from STOP_MICROPHONE to STOPPING_MICROPHONE
[00:37:42.102][D][voice_assistant:483]: State changed from STOPPING_MICROPHONE to AWAITING_RESPONSE
[00:37:42.182][D][sensor:124]: 'Sz BME680 IAQ' >> 72 IAQ
[00:37:42.183][D][sensor:124]: 'SZ BME680 CO2 Equivalent' >> 771.2 ppm
[00:37:42.194][D][sensor:124]: 'SZ BME680 Breath VOC Equivalent' >> 0.9 ppm
[00:37:42.198][D][sensor:124]: 'SZ BME680 Pressure' >> 1016.5 hPa
[00:37:42.212][D][sensor:124]: 'SZ BME680 Temperature' >> 20.22 °C
[00:37:42.212][D][sensor:124]: 'SZ BME680 Humidity' >> 47.8 %
[00:37:45.182][D][sensor:124]: 'Sz BME680 IAQ' >> 73 IAQ
[00:37:45.186][D][sensor:124]: 'SZ BME680 CO2 Equivalent' >> 781.1 ppm
[00:37:45.220][D][sensor:124]: 'SZ BME680 Breath VOC Equivalent' >> 0.9 ppm
[00:37:45.221][D][sensor:124]: 'SZ BME680 Pressure' >> 1016.5 hPa
[00:37:45.223][D][sensor:124]: 'SZ BME680 Temperature' >> 20.20 °C
[00:37:45.224][D][sensor:124]: 'SZ BME680 Humidity' >> 47.9 %
[00:37:46.305][D][dallas.temp.sensor:054]: 'temp_dallas_schlafzimmer': Got Temperature=22.500000°C
[00:37:46.306][D][sensor:124]: 'temp_dallas_schlafzimmer' >> 22.5 °C
[00:37:46.612][D][binary_sensor:048]: 'Schlafzimmerstation Bewegungsmelder' >> ON
[00:37:47.040][D][micro_wake_word:325]: Detected 'Okay Nabu' with sliding average probability is 0.59 and max probability is 0.98
[00:37:47.045][D][voice_assistant:611]: Signaling stop
[00:37:47.048][D][voice_assistant:631]: Event Type: 2
[00:37:47.051][D][voice_assistant:773]: Assist Pipeline ended
[00:37:47.052][D][voice_assistant:483]: State changed from AWAITING_RESPONSE to IDLE
[00:37:47.052][D][voice_assistant:490]: Desired state set to IDLE
[00:37:47.063][D][light:079]: 'sz_voice_assistant_leds' Setting:
[00:37:47.066][D][light:092]:   State: OFF
[00:37:47.067][D][light:153]:   Effect: 'None'
[00:37:47.067][D][light:079]: 'LED Ring' Setting:
[00:37:48.185][D][sensor:124]: 'Sz BME680 IAQ' >> 74 IAQ
[00:37:48.194][D][sensor:124]: 'SZ BME680 CO2 Equivalent' >> 791.9 ppm
[00:37:48.202][D][sensor:124]: 'SZ BME680 Breath VOC Equivalent' >> 1.0 ppm
[00:37:48.205][D][sensor:124]: 'SZ BME680 Pressure' >> 1016.5 hPa
[00:37:48.218][D][sensor:124]: 'SZ BME680 Temperature' >> 20.20 °C
[00:37:48.221][D][sensor:124]: 'SZ BME680 Humidity' >> 48.0 %
[00:37:50.545][D][micro_wake_word:325]: Detected 'Okay Nabu' with sliding average probability is 0.65 and max probability is 0.99
[00:37:50.546][D][media_player:113]: 'Schlafzimmer Media Player' - Setting
[00:37:50.546][D][media_player:117]:   Command: STOP
[00:37:50.549][D][media_player:126]:  Announcement: yes
[00:37:50.558][D][speaker_media_player:454]: State changed to ANNOUNCING
[00:37:50.559][D][speaker_media_player.pipeline:119]: Reading FLAC file type
[00:37:50.559][D][ring_buffer:034][ann_read]: Created ring buffer with size 100000
[00:37:50.654][D][speaker_media_player.pipeline:129]: Decoded audio has 1 channels, 48000 Hz sample rate, and 16 bits per sample
[00:37:50.661][D][resampler_speaker:120]: Starting
[00:37:50.666][D][ring_buffer:034]: Created ring buffer with size 9600
[00:37:50.670][D][speaker_mixer:454]: Starting
[00:37:50.674][D][ring_buffer:034][resampler]: Created ring buffer with size 48000
[00:37:50.696][D][i2s_audio.speaker:102]: Starting
[00:37:50.697][E][i2s_audio.speaker:521]: Parent I2S bus not free
[00:37:50.697][E][i2s_audio.speaker:148]: Driver failed to start; retrying in 1 second
[00:37:50.700][E][component:420]: i2s_audio.speaker set Error flag: unspecified
[00:37:50.700][D][ring_buffer:034][speaker_task]: Created ring buffer with size 12800
[00:37:50.713][D][i2s_audio.speaker:106]: Started
[00:37:50.841][D][voice_assistant:483]: State changed from IDLE to START_MICROPHONE
[00:37:50.843][D][voice_assistant:490]: Desired state set to START_PIPELINE
[00:37:50.844][D][voice_assistant:208]: Starting Microphone
[00:37:50.844][D][ring_buffer:034]: Created ring buffer with size 16384
[00:37:50.844][D][voice_assistant:483]: State changed from START_MICROPHONE to STARTING_MICROPHONE
[00:37:50.847][D][voice_assistant:483]: State changed from STARTING_MICROPHONE to START_PIPELINE
[00:37:50.866][D][voice_assistant:229]: Requesting start
[00:37:50.872][D][voice_assistant:483]: State changed from START_PIPELINE to STARTING_PIPELINE
[00:37:50.876][D][voice_assistant:505]: Client started, streaming microphone
[00:37:50.879][D][voice_assistant:483]: State changed from STARTING_PIPELINE to STREAMING_MICROPHONE
[00:37:50.880][D][voice_assistant:490]: Desired state set to STREAMING_MICROPHONE
[00:37:50.883][D][voice_assistant:631]: Event Type: 1
[00:37:50.889][D][voice_assistant:634]: Assist Pipeline running
[00:37:50.900][D][voice_assistant:631]: Event Type: 3
[00:37:50.906][D][voice_assistant:653]: STT started
[00:37:50.906][D][light:079]: 'sz_voice_assistant_leds' Setting:
[00:37:50.910][D][light:092]:   State: ON
[00:37:50.913][D][light:067]:   Brightness: 20%
[00:37:50.920][D][light:153]:   Effect: 'Waiting for Command'
[00:37:50.956][D][speaker_media_player:454]: State changed to IDLE
[00:37:50.958][D][resampler_speaker:151]: Stopped
[00:37:51.184][D][sensor:124]: 'Sz BME680 IAQ' >> 75 IAQ
[00:37:51.188][D][sensor:124]: 'SZ BME680 CO2 Equivalent' >> 809.7 ppm
[00:37:51.208][D][sensor:124]: 'SZ BME680 Breath VOC Equivalent' >> 1.0 ppm
[00:37:51.208][D][sensor:124]: 'SZ BME680 Pressure' >> 1016.5 hPa
[00:37:51.217][D][sensor:124]: 'SZ BME680 Temperature' >> 20.21 °C
[00:37:51.220][D][sensor:124]: 'SZ BME680 Humidity' >> 48.1 %
[00:37:51.591][D][binary_sensor:048]: 'Schlafzimmerstation Bewegungsmelder' >> OFF
[00:37:51.680][D][speaker_mixer:472]: Stopped
[00:37:51.697][E][component:433]: i2s_audio.speaker cleared Error flag
[00:37:51.914][D][i2s_audio.speaker:111]: Stopping
[00:37:51.918][D][i2s_audio.speaker:116]: Stopped
[00:37:51.984][D][voice_assistant:631]: Event Type: 11
[00:37:51.986][D][voice_assistant:834]: Starting STT by VAD
[00:37:51.991][D][light:079]: 'sz_voice_assistant_leds' Setting:
[00:37:51.992][D][light:067]:   Brightness: 20%
[00:37:51.992][D][light:153]:   Effect: 'Listening For Command'
[00:37:54.185][D][sensor:124]: 'Sz BME680 IAQ' >> 76 IAQ

Some tips on how to proceed to find the error would be welcome :o)

2

[E][i2s_audio.speaker:521]: Parent I2S bus not free
[E][i2s_audio.speaker:148]: Driver failed to start; retrying in 1 second

I saw somewhere that media component turns on WiFi hi performance and turning that off may help this.

It wasn’t specific to this but that i2s problem is your issue so that’s were to start. I was thinking that other issue I mentioned could help this as it will free memory which may be causing i2s problem.

If I find exactly what is is I will edit this post with it

EDIT

From here


If you experience out-of-memory issues, you can disable these optimizations by setting enable_high_performance: false in the Network component configuration.

I don’t see out of memory errors so may not related but can’t hurt

3

OK, with “enable_high_performance: false” the error messages are away. But there is still no answer from the assistant.

[19:54:49.844][D][micro_wake_word:325]: Detected 'Okay Nabu' with sliding average probability is 0.72 and max probability is 0.93
[19:54:49.844][D][voice_assistant:611]: Signaling stop
[19:54:49.849][D][voice_assistant:631]: Event Type: 2
[19:54:49.849][D][voice_assistant:773]: Assist Pipeline ended
[19:54:49.849][D][voice_assistant:483]: State changed from AWAITING_RESPONSE to IDLE
[19:54:49.849][D][voice_assistant:490]: Desired state set to IDLE
[19:54:49.865][D][light:079]: 'sz_voice_assistant_leds' Setting:
[19:54:49.865][D][light:092]:   State: OFF
[19:54:49.865][D][light:153]:   Effect: 'None'
[19:54:49.865][D][light:079]: 'LED Ring' Setting:
[19:54:51.773][D][micro_wake_word:325]: Detected 'Okay Nabu' with sliding average probability is 0.64 and max probability is 0.85
[19:54:51.776][D][media_player:113]: 'Schlafzimmer Media Player' - Setting
[19:54:51.776][D][media_player:117]:   Command: STOP
[19:54:51.776][D][media_player:126]:  Announcement: yes
[19:54:51.786][D][speaker_media_player:454]: State changed to ANNOUNCING
[19:54:51.843][D][speaker_media_player.pipeline:119]: Reading FLAC file type
[19:54:51.843][D][ring_buffer:034][ann_read]: Created ring buffer with size 100000
[19:54:51.888][D][speaker_media_player.pipeline:129]: Decoded audio has 1 channels, 48000 Hz sample rate, and 16 bits per sample
[19:54:51.902][D][resampler_speaker:120]: Starting
[19:54:51.902][D][ring_buffer:034]: Created ring buffer with size 9600
[19:54:51.902][D][speaker_mixer:454]: Starting
[19:54:51.902][D][ring_buffer:034][resampler]: Created ring buffer with size 48000
[19:54:51.927][D][i2s_audio.speaker:102]: Starting
[19:54:52.060][D][i2s_audio.speaker:106]: Started
[19:54:52.060][D][ring_buffer:034][speaker_task]: Created ring buffer with size 12800
[19:54:52.080][D][voice_assistant:483]: State changed from IDLE to START_MICROPHONE
[19:54:52.084][D][voice_assistant:490]: Desired state set to START_PIPELINE
[19:54:52.084][D][voice_assistant:208]: Starting Microphone
[19:54:52.084][D][ring_buffer:034]: Created ring buffer with size 16384
[19:54:52.084][D][voice_assistant:483]: State changed from START_MICROPHONE to STARTING_MICROPHONE
[19:54:52.113][D][voice_assistant:483]: State changed from STARTING_MICROPHONE to START_PIPELINE
[19:54:52.113][D][voice_assistant:229]: Requesting start
[19:54:52.113][D][voice_assistant:483]: State changed from START_PIPELINE to STARTING_PIPELINE
[19:54:52.113][D][voice_assistant:505]: Client started, streaming microphone
[19:54:52.117][D][voice_assistant:483]: State changed from STARTING_PIPELINE to STREAMING_MICROPHONE
[19:54:52.129][D][voice_assistant:490]: Desired state set to STREAMING_MICROPHONE
[19:54:52.129][D][voice_assistant:631]: Event Type: 1
[19:54:52.129][D][voice_assistant:634]: Assist Pipeline running
[19:54:52.138][D][voice_assistant:631]: Event Type: 3
[19:54:52.150][D][voice_assistant:653]: STT started
[19:54:52.150][D][light:079]: 'sz_voice_assistant_leds' Setting:
[19:54:52.150][D][light:092]:   State: ON
[19:54:52.160][D][light:067]:   Brightness: 20%
[19:54:52.189][D][light:153]:   Effect: 'Waiting for Command'
[19:54:52.189][D][speaker_media_player:454]: State changed to IDLE
[19:54:52.189][D][resampler_speaker:151]: Stopped
[19:54:52.919][D][speaker_mixer:472]: Stopped
[19:54:53.167][D][i2s_audio.speaker:111]: Stopping
[19:54:53.170][D][i2s_audio.speaker:116]: Stopped
[19:54:53.170][D][voice_assistant:631]: Event Type: 11
[19:54:53.170][D][voice_assistant:834]: Starting STT by VAD
[19:54:53.186][D][light:079]: 'sz_voice_assistant_leds' Setting:
[19:54:53.230][D][light:067]:   Brightness: 20%
[19:54:53.230][D][light:153]:   Effect: 'Listening For Command'
[19:54:55.297][D][voice_assistant:631]: Event Type: 12
[19:54:55.297][D][voice_assistant:838]: STT by VAD end
[19:54:55.300][D][voice_assistant:483]: State changed from STREAMING_MICROPHONE to STOP_MICROPHONE
[19:54:55.300][D][voice_assistant:490]: Desired state set to AWAITING_RESPONSE
[19:54:55.300][D][voice_assistant:483]: State changed from STOP_MICROPHONE to STOPPING_MICROPHONE
[19:54:55.300][D][voice_assistant:483]: State changed from STOPPING_MICROPHONE to AWAITING_RESPONSE
4

I detected “Speech-to-Phrase” is not starting.
I uninstalled it, restarted the Raspberry Pi, reinstalled it, and restarted Home Assistant.
The app seems to be stuck at startup…

Log of Speech-to-Phrase (always repeating in a loop):

AttributeError: 'NoneType' object has no attribute 'strip'
[20:44:04] INFO: Service exited with code 1 (by signal 0)
2026-04-02 20:44:07.391596038 [W:onnxruntime:Default, device_discovery.cc:211 DiscoverDevicesForPlatform] GPU device discovery failed: device_discovery.cc:91 ReadFileContents Failed to open file: "/sys/class/drm/card1/device/vendor"
Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "/usr/src/.venv/lib/python3.11/site-packages/speech_to_phrase/__main__.py", line 191, in <module>
    asyncio.run(main())
  File "/usr/lib/python3.11/asyncio/runners.py", line 190, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.11/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.11/asyncio/base_events.py", line 653, in run_until_complete
    return future.result()
           ^^^^^^^^^^^^^^^
  File "/usr/src/.venv/lib/python3.11/site-packages/speech_to_phrase/__main__.py", line 96, in main
    await _retrain_once(state, force_retrain=True)
  File "/usr/src/.venv/lib/python3.11/site-packages/speech_to_phrase/__main__.py", line 131, in _retrain_once
    hass_info = await get_hass_info(
                ^^^^^^^^^^^^^^^^^^^^
  File "/usr/src/.venv/lib/python3.11/site-packages/speech_to_phrase/hass_api.py", line 413, in get_hass_info
    names=[name.strip() for name in names],
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/src/.venv/lib/python3.11/site-packages/speech_to_phrase/hass_api.py", line 413, in <listcomp>
    names=[name.strip() for name in names],
           ^^^^^^^^^^
5

Yes there is a crash loop with speech to phrase. Issue has been opened on github and best to stop it/uninstall it until a fix is done. Someone did post a workaround.

6

The Wyoming Protocol is also not detecting speech to phrase any more. Ok, I will uninstall it and wait…

7

Speech to phrase looks at your areas and entities and trains on them or something

Anyway one of your names or areas has a “ “ or some illegal character that cannot be recognized. Check beginning and end of entities names and area names for “ “. If it worked before and you added device recently start there

8

In my texts (conversations), I use German umlauts like “ü, ä, ö”, for example in the words “Lautstärke” (volume), “Tür” (door), or “Rolläden” (shutters). I hope that Speech-to-Phrase won’t have any problems with that.

In the areas, I have “Büro” (office) and “Küche” (kitchen) with special characters, which worked for weeks, but there are no quotation marks in the names. Spaces are part of many entity names (such as “Energy Total”) but not part of the entity ID.

There is a name for an entity “Extra LED’s Camera” with “'” in it, but it has worked for a long time and is not used in connection with Speech-To-Phrase.

…As a first step, I’ll change the few area names and the entity name. I hope I won’t have to rewrite the entire conversation.

9

…Ok the easy way is not working. I have to wait for a fix or better information about the position of the faulty name. If I have to remove every ü, ä or ö from every entity name, then I can make everything new…

10

It will say exactly in speech 2 phrase log.
Not sure u go in access in HAOS.

11

You mean the protocol ?

  File "/usr/src/.venv/lib/python3.11/site-packages/speech_to_phrase/hass_api.py", line 413, in <listcomp>
    names=[name.strip() for name in names],
           ^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'strip'
[13:13:03] INFO: e[32mService exited with code 1 (by signal 0)e[0m
e[0;93m2026-04-03 13:13:06.572010308 [W:onnxruntime:Default, device_discovery.cc:211 DiscoverDevicesForPlatform] GPU device discovery failed: device_discovery.cc:91 ReadFileContents Failed to open file: "/sys/class/drm/card1/device/vendor"e[m
Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "/usr/src/.venv/lib/python3.11/site-packages/speech_to_phrase/__main__.py", line 191, in <module>
    asyncio.run(main())
  File "/usr/lib/python3.11/asyncio/runners.py", line 190, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.11/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.11/asyncio/base_events.py", line 653, in run_until_complete
    return future.result()
           ^^^^^^^^^^^^^^^
  File "/usr/src/.venv/lib/python3.11/site-packages/speech_to_phrase/__main__.py", line 96, in main
    await _retrain_once(state, force_retrain=True)
  File "/usr/src/.venv/lib/python3.11/site-packages/speech_to_phrase/__main__.py", line 131, in _retrain_once
    hass_info = await get_hass_info(
                ^^^^^^^^^^^^^^^^^^^^
  File "/usr/src/.venv/lib/python3.11/site-packages/speech_to_phrase/hass_api.py", line 413, in get_hass_info
    names=[name.strip() for name in names],
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/src/.venv/lib/python3.11/site-packages/speech_to_phrase/hass_api.py", line 413, in <listcomp>
    names=[name.strip() for name in names],
           ^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'strip'
[13:13:07] INFO: e[32mService exited with code 1 (by signal 0)e[0m
e[0;93m2026-04-03 13:13:10.043941914 [W:onnxruntime:Default, device_discovery.cc:211 DiscoverDevicesForPlatform] GPU device discovery failed: device_discovery.cc:91 ReadFileContents Failed to open file: "/sys/class/drm/card1/device/vendor"e[m
Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main

or where can I find the speech 2 phrase log?

12

FYI

13

Great, the bug fix for Speech-To-Phrase is here (version 1.4.3). Installed it and it’s working again :slightly_smiling_face: