TTS Playback delay

I’ve setup Snips using a Raspberry Pi 3 Model B+

When it plays back tts, the first couple of seconds of the speech is not heard.

For example, if I tell it to say “Do you have something you’d like to share?” it only speaks “something you’d like to share?”.

Any pointers?

can you try using snips-watch -vvv and paste some of the output of what happens when you ask it to talk here ?

Of course. Here is what I get

pi@raspberrypi:~ $ snips-watch -vvv
[23:16:25] Watching on localhost:1883 (MQTT)
[23:16:25] [Dialogue] was asked to start a session
[23:16:29] [Dialogue] was asked to start a session
[23:16:29] [Asr] was asked to stop listening on site default
[23:16:29] [Hotword] was asked to toggle itself 'off' on site default
[23:16:29] [Dialogue] session with id '9d24d009-c8f3-4c41-835f-689bc3dfc161' was queued on site default
[23:16:29] [Dialogue] session with id '9d24d009-c8f3-4c41-835f-689bc3dfc161' was started on site default
[23:16:29] [Tts] was asked to say "Do you have something you’d like to share?"
[23:16:29] [AudioServer] was asked to play a wav of 85.4 kB with id '26ba9ad9-4c70-4a4f-81e0-cf25f809a54c' on site default
[23:16:32] [AudioServer] finished playing wav with id '26ba9ad9-4c70-4a4f-81e0-cf25f809a54c'
[23:16:32] [Tts] finished speaking with id '2327191f-362e-424b-bc8e-eb47eb9245bf'
[23:16:32] [AudioServer] was asked to play a wav of 41.1 kB with id 'd6fb0dab-6ffa-485b-a1f3-1f7ab000ccde' on site default
[23:16:33] [AudioServer] finished playing wav with id 'd6fb0dab-6ffa-485b-a1f3-1f7ab000ccde'
[23:16:33] [Asr] was asked to listen on site default
[23:16:38] [Asr] captured text "" in 4.0s
[23:16:38] [Asr] was asked to stop listening on site default
[23:16:38] [AudioServer] was asked to play a wav of 93.1 kB with id '015d8f52-e29c-45d3-8f39-9e9ce59df4c0' on site default
[23:16:39] [AudioServer] finished playing wav with id '015d8f52-e29c-45d3-8f39-9e9ce59df4c0'
[23:16:39] [AudioServer] was asked to play a wav of 61.1 kB with id '556cdddb-c33c-4ea1-a930-2087d050c215' on site default
[23:16:39] [AudioServer] finished playing wav with id '556cdddb-c33c-4ea1-a930-2087d050c215'

try snips-tts --help
this will output some info, and the very last item is the output directory where the wave files go…
open that directory and find the (i think it will be something like) 26ba9ad9-4c70-4a4f-81e0-cf25f809a54c.wav and try to use aplay 26ba9ad9-4c70-4a4f-81e0-cf25f809a54c.wav and just hear what it has
Im trying to ascertain if the wave file contains the whole sentence, or if audio-server is the problem

The last line says:

-u, --user-dir Path to a writable persistent work directory [default: /var/lib/snips/]

But the only thing in that folder is the skills folder:

pi@raspberrypi:/var/lib/snips $ ls /var/lib/snips
skills

Hello,

I had similar problem when using an USB sound card with digital output.
With the analog output it worked.

Thanks. I switched from using HDMI to the audio out and it worked!