NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
PyPI · #2992 most downloaded on PyPI
Agent Framework plugin for services using Deepgram's API.
Last release 4 days ago
01 Oct 2026
Ships fairly regularly
a new release about every 8 days
Rarely documented
notes for 6 of the last 60 stable releases
Nothing withdrawn
no release was ever pulled
3 years old
167 releases · first in 2024
support multilingual with Nova-3 model - #1736 (@jeradf)
Added optional parameter to opt out from deepgrams model improvement plan - #1713 (@MatthiasGruba)
One column per month.
add nova-3-medical to stt models - #1657 (@jayeshp19)
add nova-3-medical to stt models - #1657 (@jayeshp19)
Add string type support to model parameter - #1657 (@jayeshp19)
support numerals deepgram stt - #1667 (@jayeshp19)
use streaming AudioDecoder to handle compressed encoding - #1584 (@davidzhao)
added a tts.prewarm method to start the connection pool early. - #1587 (@davidzhao)
update pool configuration for deepgram and cartesia - #1605 (@jayeshp19)
set mex session duration to 1 hour in deepgram connection pool - #1582 (@jayeshp19)
updated livekit-agent reference to <1.0 - #1607 (@davidzhao)
fix(deepgram): fix STT keyterm parameter - #1535 (@wdhwg001)
fix(deepgram): fix STT keyterm parameter - #1535 (@wdhwg001)
use connection pool for deepgram tts - #1523 (@jayeshp19)
remove update options from tts synthesis stream - #1546 (@jayeshp19)
deepgram: support for Nova-3 keyterms - #1484 (@davidzhao)
chore(Deepgram STT): add nova-3 model to type literal - #1464 (@chasemcdo)
improved TTFB metrics for streaming TTS - #1431 (@davidzhao)
fix: Ensure STT exceptions are being propagated - #1291 (@davidzhao)
added streaming audio decoder for compressed audio. - #1236 (@davidzhao)
added streaming audio decoder for compressed audio. - #1236 (@davidzhao)
Support Deepgram TTS - #1201 (@jayeshp19)
enable deepgram filler words by default to improve end of turn accuracy - #1190 (@davidzhao)
fix release - #1176 (@theomonnom)
add PeriodicCollector utility for metrics - #1094 (@davidzhao)
fix Deepgram missing first word, disabled energy filter by default - #1090 (@davidzhao)
stt: reduce bandwidth usage by reducing sample_rate to 16khz - #920 (@theomonnom)
stt: reduce bandwidth usage by reducing sample_rate to 16khz - #920 (@theomonnom)
deepgram: send finalize each time we stop sending audio - #1004 (@theomonnom)
pipelineagent: expose timing metrics & api errors wip - #957 (@theomonnom)
expose usage metrics - #984 (@theomonnom)
accepts parameter profanity_filter - #811 (@jebjebs)
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Your coding agent can read these notes before it upgrades. Set up the MCP server →