Overview
the first time I ask voice assistant a question the sound does not play. When I ask a second time shortly after, I do hear his answer. When I now wait a few minutes and ask something again, again the sound does not play.
My Setup
I have a conference microphone with speaker ( Anker PowerConf S330 ) connected directly to my RasPi 5 which runs Home Assistant OS.
I use:
- Assist Microphone Addon (for local microphone)
- Openwakeword with a custom wake word
- ChatGPT as conversation agent
- GroqCloud Whisper as SST, but also tested with Google AI STT
- OpenAI TTS, but also tested with piper, google translate and Google AI TTS
My Observations
I think at some point this used to work. At the beginnings of the Voice Assistant. My whole setup.
I think it broke at sometime spring last year.
What happens:
The first time I ask a question after a while I hear no answer. In the debug logs I can see no errors. The agent did answer I just could not hear it. the chime at the start and end of my question does play, so I doubt itβs an audio issue.
When I ask shortly something again the sound does play correctly.
I tried all kinds of different setups but everything behaves the same.
I tried different SST, TTS and agents.
While the Debug Assistant does show the text-to-speech answer it may be worth noting that the βPlay Audioβ button always returns βError playing audioβ. But it does that in every case, even when I did hear the message being played.
Logs
Here is a log of the first question that plays NO SOUND:
Unsuccesful log
stage: done
run:
pipeline: 01kfdh20h4jbqf25cgb051jnfx
language: de
conversation_id: 01KHGKCGN4FNETWEAWJ740TGS9
satellite_id: assist_satellite.assist_microphone
tts_output:
token: TVW4A3tiBWHA8E3axq28EQ.wav
url: /api/tts_proxy/TVW4A3tiBWHA8E3axq28EQ.wav
mime_type: audio/x-wav
stream_response: true
events:
-
type: run-start
data:
pipeline: 01kfdh20h4jbqf25cgb051jnfx
language: de
conversation_id: 01KHGKCGN4FNETWEAWJ740TGS9
satellite_id: assist_satellite.assist_microphone
tts_output:
token: TVW4A3tiBWHA8E3axq28EQ.wav
url: /api/tts_proxy/TVW4A3tiBWHA8E3axq28EQ.wav
mime_type: audio/x-wav
stream_response: true
timestamp: β2026-02-15T12:11:11.911870+00:00β -
type: wake_word-start
data:
entity_id: wake_word.openwakeword
metadata:
format: wav
codec: pcm
bit_rate: 16
sample_rate: 16000
channel: 1
timeout: 0
timestamp: β2026-02-15T12:11:11.911986+00:00β -
type: wake_word-end
data:
wake_word_output:
wake_word_id: Mister_Chat_G_P_T_
wake_word_phrase: Mister Chat G P T
timestamp: 667780
timestamp: β2026-02-15T12:22:19.815337+00:00β -
type: stt-start
data:
engine: stt.groqcloud_whisper
metadata:
language: de
format: wav
codec: pcm
bit_rate: 16
sample_rate: 16000
channel: 1
timestamp: β2026-02-15T12:22:19.816498+00:00β -
type: stt-vad-start
data:
timestamp: 668200
timestamp: β2026-02-15T12:22:20.882494+00:00β -
type: stt-vad-end
data:
timestamp: 671020
timestamp: β2026-02-15T12:22:23.661177+00:00β -
type: stt-end
data:
stt_output:
text: " How is the weather today?"
timestamp: β2026-02-15T12:22:24.001334+00:00β -
type: intent-start
data:
engine: conversation.chatgpt
language: de
intent_input: " How is the weather today?"
conversation_id: 01KHGKCGN4FNETWEAWJ740TGS9
device_id: cedf41be5cb823e8278b57a3973ad448
satellite_id: assist_satellite.assist_microphone
prefer_local_intents: true
timestamp: β2026-02-15T12:22:24.001629+00:00β -
type: intent-progress
data:
chat_log_delta:
role: assistant
timestamp: β2026-02-15T12:22:26.734969+00:00β -
type: intent-progress
data:
chat_log_delta:
tool_calls:
- tool_name: WeatherForecast
tool_args: {}
id: call_vQ4HBZ01AqeM1UpBSftW4eYn
external: false
timestamp: β2026-02-15T12:22:27.168800+00:00β -
type: intent-progress
data:
chat_log_delta:
role: tool_result
agent_id: conversation.chatgpt
tool_call_id: call_vQ4HBZ01AqeM1UpBSftW4eYn
tool_name: WeatherForecast
tool_result:
speech:
plain:
speech: |-
The weather summary is: {
βforecastβ: [{'condition': 'partlycloudy', 'precipitation_probability': 0.0, 'datetime': 'Sunday', 'wind_bearing': 152.0, 'uv_index': 1.6, 'temperature': -0.2, 'templow': -3.1, 'wind_gust_speed': 43.9, 'wind_speed': 23.8, 'precipitation': 0.0, 'humidity': 67} , {'condition': 'snowy', 'precipitation_probability': 87.1, 'datetime': 'Monday', 'wind_bearing': 131.0, 'uv_index': 1.5, 'temperature': -1.9, 'templow': -3.8, 'wind_gust_speed': 45.7, 'wind_speed': 24.8, 'precipitation': 2.5, 'humidity': 83} , {'condition': 'cloudy', 'precipitation_probability': 45.3, 'datetime': 'Tuesday', 'wind_bearing': 165.0, 'uv_index': 1.6, 'temperature': 2.3, 'templow': -1.8, 'wind_gust_speed': 19.4, 'wind_speed': 10.1, 'precipitation': 2.4, 'humidity': 97} , {'condition': 'cloudy', 'precipitation_probability': 17.6, 'datetime': 'Wednesday', 'wind_bearing': 333.4, 'temperature': -1.2, 'templow': -3.8, 'wind_speed': 12.2, 'precipitation': 0.0, 'humidity': 76} , {'condition': 'cloudy', 'precipitation_probability': 3.9, 'datetime': 'Thursday', 'wind_bearing': 66.7, 'temperature': -1.4, 'templow': -6.5, 'wind_speed': 7.9, 'precipitation': 0.0, 'humidity': 77} , {'condition': 'cloudy', 'precipitation_probability': 23.5, 'datetime': 'Friday', 'wind_bearing': 203.7, 'temperature': 1.6, 'templow': -7.9, 'wind_speed': 14.8, 'precipitation': 0.0, 'humidity': 70} ] } extra_data: null response_type: action_done data: targets: [] success: [] failed: [] created: "2026-02-15T12:22:27.187732+00:00"timestamp: β2026-02-15T12:22:27.187877+00:00β
-
type: intent-progress
data:
chat_log_delta:
role: assistant
timestamp: β2026-02-15T12:22:28.677860+00:00β -
type: intent-progress
data:
chat_log_delta:
content: Todayβs
timestamp: β2026-02-15T12:22:28.678488+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " weather"
timestamp: β2026-02-15T12:22:28.710638+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " is"
timestamp: β2026-02-15T12:22:28.712253+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " partly"
timestamp: β2026-02-15T12:22:28.776324+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " cloudy"
timestamp: β2026-02-15T12:22:28.777179+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " with"
timestamp: β2026-02-15T12:22:28.842759+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " a"
timestamp: β2026-02-15T12:22:28.843882+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " temperature"
timestamp: β2026-02-15T12:22:28.968605+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " of"
timestamp: β2026-02-15T12:22:28.969435+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " minus"
timestamp: β2026-02-15T12:22:29.036708+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " point"
timestamp: β2026-02-15T12:22:29.037411+00:00β -
type: intent-progress
data:
tts_start_streaming: true
timestamp: β2026-02-15T12:22:29.037444+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " two"
timestamp: β2026-02-15T12:22:29.086249+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " degrees"
timestamp: β2026-02-15T12:22:29.087316+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " Celsius"
timestamp: β2026-02-15T12:22:29.110682+00:00β -
type: intent-progress
data:
chat_log_delta:
content: .
timestamp: β2026-02-15T12:22:29.111443+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " The"
timestamp: β2026-02-15T12:22:29.139380+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " humidity"
timestamp: β2026-02-15T12:22:29.140297+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " is"
timestamp: β2026-02-15T12:22:29.191290+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " at"
timestamp: β2026-02-15T12:22:29.192365+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " sixty"
timestamp: β2026-02-15T12:22:29.232208+00:00β -
type: intent-progress
data:
chat_log_delta:
content: β-sevenβ
timestamp: β2026-02-15T12:22:29.232846+00:00β -
type: intent-progress
data:
chat_log_delta:
content: " percent"
timestamp: β2026-02-15T12:22:29.247804+00:00β -
type: intent-progress
data:
chat_log_delta:
content: .
timestamp: β2026-02-15T12:22:29.251908+00:00β -
type: intent-end
data:
processed_locally: false
intent_output:
response:
speech:
plain:
speech: >-
Todayβs weather is partly cloudy with a temperature of minus
point two degrees Celsius. The humidity is at sixty-seven
percent.
extra_data: null
card: {}
language: de
response_type: action_done
data:
targets:
success:
failed:
conversation_id: 01KHGKCGN4FNETWEAWJ740TGS9
continue_conversation: false
timestamp: β2026-02-15T12:22:29.287456+00:00β -
type: tts-start
data:
engine: tts.openai_tts_onyx
language: de
voice: null
tts_input: >-
Todayβs weather is partly cloudy with a temperature of minus point two
degrees Celsius. The humidity is at sixty-seven percent.
acknowledge_override: false
timestamp: β2026-02-15T12:22:29.287516+00:00β -
type: tts-end
data:
tts_output:
media_id: media-source://tts/-stream-/TVW4A3tiBWHA8E3axq28EQ.wav
token: TVW4A3tiBWHA8E3axq28EQ.wav
url: /api/tts_proxy/TVW4A3tiBWHA8E3axq28EQ.wav
mime_type: audio/x-wav
timestamp: β2026-02-15T12:22:29.288244+00:00β -
type: run-end
data: null
timestamp: β2026-02-15T12:22:29.288288+00:00β
started: 2026-02-15T12:11:11.911Z
wake_word:
entity_id: wake_word.openwakeword
metadata:
format: wav
codec: pcm
bit_rate: 16
sample_rate: 16000
channel: 1
timeout: 0
done: true
wake_word_output:
wake_word_id: Mister_Chat_G_P_T_
wake_word_phrase: Mister Chat G P T
timestamp: 667780
stt:
engine: stt.groqcloud_whisper
metadata:
language: de
format: wav
codec: pcm
bit_rate: 16
sample_rate: 16000
channel: 1
done: true
stt_output:
text: " How is the weather today?"
intent:
engine: conversation.chatgpt
language: de
intent_input: " How is the weather today?"
conversation_id: 01KHGKCGN4FNETWEAWJ740TGS9
device_id: cedf41be5cb823e8278b57a3973ad448
satellite_id: assist_satellite.assist_microphone
prefer_local_intents: true
done: true
processed_locally: false
intent_output:
response:
speech:
plain:
speech: >-
Todayβs weather is partly cloudy with a temperature of minus point
two degrees Celsius. The humidity is at sixty-seven percent.
extra_data: null
card: {}
language: de
response_type: action_done
data:
targets:
success:
failed:
conversation_id: 01KHGKCGN4FNETWEAWJ740TGS9
continue_conversation: false
tts:
engine: tts.openai_tts_onyx
language: de
voice: null
tts_input: >-
Todayβs weather is partly cloudy with a temperature of minus point two
degrees Celsius. The humidity is at sixty-seven percent.
acknowledge_override: false
done: true
tts_output:
media_id: media-source://tts/-stream-/TVW4A3tiBWHA8E3axq28EQ.wav
token: TVW4A3tiBWHA8E3axq28EQ.wav
url: /api/tts_proxy/TVW4A3tiBWHA8E3axq28EQ.wav
mime_type: audio/x-wav
finished: 2026-02-15T12:22:29.288Z
And here is the log of the successful run shortly after
Successful
stage: done
run:
pipeline: 01kfdh20h4jbqf25cgb051jnfx
language: de
conversation_id: 01KHGKCGN4FNETWEAWJ740TGS9
satellite_id: assist_satellite.assist_microphone
tts_output:
token: pxTx_MsZnvJWXN-prGwvzw.wav
url: /api/tts_proxy/pxTx_MsZnvJWXN-prGwvzw.wav
mime_type: audio/x-wav
stream_response: true
events:
- type: run-start
data:
pipeline: 01kfdh20h4jbqf25cgb051jnfx
language: de
conversation_id: 01KHGKCGN4FNETWEAWJ740TGS9
satellite_id: assist_satellite.assist_microphone
tts_output:
token: pxTx_MsZnvJWXN-prGwvzw.wav
url: /api/tts_proxy/pxTx_MsZnvJWXN-prGwvzw.wav
mime_type: audio/x-wav
stream_response: true
timestamp: β2026-02-15T12:22:29.296949+00:00β - type: wake_word-start
data:
entity_id: wake_word.openwakeword
metadata:
format: wav
codec: pcm
bit_rate: 16
sample_rate: 16000
channel: 1
timeout: 0
timestamp: β2026-02-15T12:22:29.297046+00:00β - type: wake_word-end
data:
wake_word_output:
wake_word_id: Mister_Chat_G_P_T_
wake_word_phrase: Mister Chat G P T
timestamp: 12180
timestamp: β2026-02-15T12:22:41.451621+00:00β - type: stt-start
data:
engine: stt.groqcloud_whisper
metadata:
language: de
format: wav
codec: pcm
bit_rate: 16
sample_rate: 16000
channel: 1
timestamp: β2026-02-15T12:22:41.454288+00:00β - type: stt-vad-start
data:
timestamp: 12630
timestamp: β2026-02-15T12:22:42.514010+00:00β - type: stt-vad-end
data:
timestamp: 16140
timestamp: β2026-02-15T12:22:46.085965+00:00β - type: stt-end
data:
stt_output:
text: " Kannst du deine letzte Antwort wiederholen?"
timestamp: β2026-02-15T12:22:46.437168+00:00β - type: intent-start
data:
engine: conversation.chatgpt
language: de
intent_input: " Kannst du deine letzte Antwort wiederholen?"
conversation_id: 01KHGKCGN4FNETWEAWJ740TGS9
device_id: cedf41be5cb823e8278b57a3973ad448
satellite_id: assist_satellite.assist_microphone
prefer_local_intents: true
timestamp: β2026-02-15T12:22:46.437465+00:00β - type: intent-progress
data:
chat_log_delta:
role: assistant
timestamp: β2026-02-15T12:22:48.100600+00:00β - type: intent-progress
data:
chat_log_delta:
content: Heute
timestamp: β2026-02-15T12:22:48.101833+00:00β - type: intent-progress
data:
chat_log_delta:
content: " ist"
timestamp: β2026-02-15T12:22:48.120665+00:00β - type: intent-progress
data:
chat_log_delta:
content: " das"
timestamp: β2026-02-15T12:22:48.121325+00:00β - type: intent-progress
data:
chat_log_delta:
content: " Wetter"
timestamp: β2026-02-15T12:22:48.145384+00:00β - type: intent-progress
data:
chat_log_delta:
content: " teilweise"
timestamp: β2026-02-15T12:22:48.146544+00:00β - type: intent-progress
data:
chat_log_delta:
content: " bew"
timestamp: β2026-02-15T12:22:48.167360+00:00β - type: intent-progress
data:
chat_log_delta:
content: ΓΆl
timestamp: β2026-02-15T12:22:48.168593+00:00β - type: intent-progress
data:
chat_log_delta:
content: kt
timestamp: β2026-02-15T12:22:48.189818+00:00β - type: intent-progress
data:
chat_log_delta:
content: " bei"
timestamp: β2026-02-15T12:22:48.191109+00:00β - type: intent-progress
data:
chat_log_delta:
content: " einer"
timestamp: β2026-02-15T12:22:48.208808+00:00β - type: intent-progress
data:
chat_log_delta:
content: " Temperatur"
timestamp: β2026-02-15T12:22:48.209825+00:00β - type: intent-progress
data:
chat_log_delta:
content: " von"
timestamp: β2026-02-15T12:22:48.250298+00:00β - type: intent-progress
data:
tts_start_streaming: true
timestamp: β2026-02-15T12:22:48.250383+00:00β - type: intent-progress
data:
chat_log_delta:
content: " minus"
timestamp: β2026-02-15T12:22:48.253626+00:00β - type: intent-progress
data:
chat_log_delta:
content: " zwei"
timestamp: β2026-02-15T12:22:48.284712+00:00β - type: intent-progress
data:
chat_log_delta:
content: " Grad"
timestamp: β2026-02-15T12:22:48.285754+00:00β - type: intent-progress
data:
chat_log_delta:
content: " Celsius"
timestamp: β2026-02-15T12:22:48.311058+00:00β - type: intent-progress
data:
chat_log_delta:
content: .
timestamp: β2026-02-15T12:22:48.312175+00:00β - type: intent-progress
data:
chat_log_delta:
content: " Die"
timestamp: β2026-02-15T12:22:48.329729+00:00β - type: intent-progress
data:
chat_log_delta:
content: " Luft"
timestamp: β2026-02-15T12:22:48.330993+00:00β - type: intent-progress
data:
chat_log_delta:
content: fe
timestamp: β2026-02-15T12:22:48.353970+00:00β - type: intent-progress
data:
chat_log_delta:
content: uchtigkeit
timestamp: β2026-02-15T12:22:48.355589+00:00β - type: intent-progress
data:
chat_log_delta:
content: " betrΓ€gt"
timestamp: β2026-02-15T12:22:48.376080+00:00β - type: intent-progress
data:
chat_log_delta:
content: " sechs"
timestamp: β2026-02-15T12:22:48.376853+00:00β - type: intent-progress
data:
chat_log_delta:
content: und
timestamp: β2026-02-15T12:22:48.423794+00:00β - type: intent-progress
data:
chat_log_delta:
content: se
timestamp: β2026-02-15T12:22:48.424879+00:00β - type: intent-progress
data:
chat_log_delta:
content: ch
timestamp: β2026-02-15T12:22:48.448980+00:00β - type: intent-progress
data:
chat_log_delta:
content: zig
timestamp: β2026-02-15T12:22:48.450444+00:00β - type: intent-progress
data:
chat_log_delta:
content: " Prozent"
timestamp: β2026-02-15T12:22:48.459031+00:00β - type: intent-progress
data:
chat_log_delta:
content: .
timestamp: β2026-02-15T12:22:48.462350+00:00β - type: intent-end
data:
processed_locally: false
intent_output:
response:
speech:
plain:
speech: >-
Heute ist das Wetter teilweise bewΓΆlkt bei einer Temperatur von
minus zwei Grad Celsius. Die Luftfeuchtigkeit betrΓ€gt
sechsundsechzig Prozent.
extra_data: null
card: {}
language: de
response_type: action_done
data:
targets:
success:
failed:
conversation_id: 01KHGKCGN4FNETWEAWJ740TGS9
continue_conversation: false
timestamp: β2026-02-15T12:22:48.502026+00:00β - type: tts-start
data:
engine: tts.openai_tts_onyx
language: de
voice: null
tts_input: >-
Heute ist das Wetter teilweise bewΓΆlkt bei einer Temperatur von minus
zwei Grad Celsius. Die Luftfeuchtigkeit betrΓ€gt sechsundsechzig Prozent.
acknowledge_override: false
timestamp: β2026-02-15T12:22:48.502082+00:00β - type: tts-end
data:
tts_output:
media_id: media-source://tts/-stream-/pxTx_MsZnvJWXN-prGwvzw.wav
token: pxTx_MsZnvJWXN-prGwvzw.wav
url: /api/tts_proxy/pxTx_MsZnvJWXN-prGwvzw.wav
mime_type: audio/x-wav
timestamp: β2026-02-15T12:22:48.503459+00:00β - type: run-end
data: null
timestamp: β2026-02-15T12:22:48.503500+00:00β
started: 2026-02-15T12:22:29.296Z
wake_word:
entity_id: wake_word.openwakeword
metadata:
format: wav
codec: pcm
bit_rate: 16
sample_rate: 16000
channel: 1
timeout: 0
done: true
wake_word_output:
wake_word_id: Mister_Chat_G_P_T_
wake_word_phrase: Mister Chat G P T
timestamp: 12180
stt:
engine: stt.groqcloud_whisper
metadata:
language: de
format: wav
codec: pcm
bit_rate: 16
sample_rate: 16000
channel: 1
done: true
stt_output:
text: " Kannst du deine letzte Antwort wiederholen?"
intent:
engine: conversation.chatgpt
language: de
intent_input: " Kannst du deine letzte Antwort wiederholen?"
conversation_id: 01KHGKCGN4FNETWEAWJ740TGS9
device_id: cedf41be5cb823e8278b57a3973ad448
satellite_id: assist_satellite.assist_microphone
prefer_local_intents: true
done: true
processed_locally: false
intent_output:
response:
speech:
plain:
speech: >-
Heute ist das Wetter teilweise bewΓΆlkt bei einer Temperatur von
minus zwei Grad Celsius. Die Luftfeuchtigkeit betrΓ€gt
sechsundsechzig Prozent.
extra_data: null
card: {}
language: de
response_type: action_done
data:
targets:
success:
failed:
conversation_id: 01KHGKCGN4FNETWEAWJ740TGS9
continue_conversation: false
tts:
engine: tts.openai_tts_onyx
language: de
voice: null
tts_input: >-
Heute ist das Wetter teilweise bewΓΆlkt bei einer Temperatur von minus zwei
Grad Celsius. Die Luftfeuchtigkeit betrΓ€gt sechsundsechzig Prozent.
acknowledge_override: false
done: true
tts_output:
media_id: media-source://tts/-stream-/pxTx_MsZnvJWXN-prGwvzw.wav
token: pxTx_MsZnvJWXN-prGwvzw.wav
url: /api/tts_proxy/pxTx_MsZnvJWXN-prGwvzw.wav
mime_type: audio/x-wav
finished: 2026-02-15T12:22:48.503Z
I couldnβt find anything else relevant in any other logs.