Error in _stt_task ... Why am I getting this err w...
# ask-ai
h
Error in _stt_task ... Why am I getting this err when I swtched to speechmatics. stt = speechmatics.STT( language="hi" )
t
Thanks for reaching out, Abhishek. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.
Thanks — I’m responding to your question about the
_stt_task
error after switching to Speechmatics and the snippet you shared. I need the exact error message and stack trace to diagnose this precisely. Could you paste the full error output? Meanwhile, please confirm these quick checks: • You installed the Speechmatics plugin:
uv add "livekit-agents[speechmatics]~=1.3"
. •
SPEECHMATICS_API_KEY
is set in your environment (e.g.,
.env
) and valid. • The language code
hi
(Hindi) is supported by your Speechmatics model/account. If you share the error text I’ll pinpoint whether it’s an unsupported language, missing/invalid API key, model configuration, or a runtime
_stt_task
exception and give the exact next steps. Sources: Speechmatics STT plugin guide | LiveKit docs
h
$ pip list | grep livekit livekit 1.0.23 livekit-agents 1.3.9 livekit-api 1.1.0 livekit-blingfire 1.0.0 livekit-plugins-cartesia 1.3.9 livekit-plugins-deepgram 1.3.9 livekit-plugins-elevenlabs 1.3.9 livekit-plugins-groq 1.3.9 livekit-plugins-langchain 1.3.9 livekit-plugins-noise-cancellation 0.2.5 livekit-plugins-openai 1.3.9 livekit-plugins-sarvam 1.3.9 livekit-plugins-silero 1.3.9 livekit-plugins-speechmatics 1.3.9 livekit-plugins-turn-detector 1.3.9 livekit-protocol 1.1.1 anything wrong with packages that I have?
t
Thanks — I’m responding to your question about whether anything is wrong with the packages you listed. Your installed packages and versions look fine and include the Speechmatics plugin (
livekit-plugins-speechmatics 1.3.9
), so there’s no obvious package-version issue. Please check these next items so I can narrow the cause of the
_stt_task
error: • Confirm
SPEECHMATICS_API_KEY
is set and correct in your environment (
.env
or process env). • Share the full error message and stack trace from
_stt_task
— that’s the fastest way to pinpoint whether it’s an unsupported language, auth failure, or runtime bug. • If possible, try a known-supported language (e.g.,
en
) to see if the same error occurs. Share the error output and results of the
en
test and I’ll guide the exact fix.