Blog post misstates when ASR interim transcripts are sent to the LLM
Maintainers usually reply within 1 day
Nobody has claimed this yet.
Assessment
- Difficulty
- 1/5
- Estimated time
- Under an hour
- Newbie friendliness
- 78/100
- Issue type
- Documentation
- Clarity
- Clearly specified
- Activity status
- Active
- Tech stack
- python
- Domain
- ai, documentation
Research direction
Read the linked blog post, “Voice Agent Architecture: STT-LLM-TTS Pipelines Explained,” and compare its description of interim ASR transcripts with the behavior stated in this issue. Update the wording to explain that interim text is held in-process and sent to the LLM only after the final transcript arrives or end of turn is detected; done when the post no longer implies the LLM starts on interim transcripts.
Written by the indexing model from the issue text.
Description
The blog at https://livekit.com/blog/voice-agent-architecture-stt-llm-tts-pipelines-explained makes it seem like the ASR interim transcripts are streamed to the LLM, and the LLM starts running before the final transcription arrives, this is False. The ASR does stream, but the interim transcripts are held in-process, and when the final transcript arrives or when the end of turn is detected only then is the held text sent to the LLM.
- Dominant language
- Python
- Stars
- 14.6k
- Forks
- 3.9k
- Avg merge
- 2d 5h
- Merged PRs (30d)
- 232
Getting set up
This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from livekit/agents
-
mistralai: expose reasoning_effort on the LLMPossibly taken @muzman123 claimed this today. Openenhancement
Difficulty 2/5 1-3 hours Newbie friendliness 66/100
Maintainers usually reply within 1 day
-
ci: unit tests pin Python 3.12 while the package advertises 3.14 supportPossibly taken @huong8373tt-beep claimed this 1 day ago. Open
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
livekit/agents#8231 · 3 comments ·
Maintainers usually reply within 1 day
-
Assistant markup is stripped from chat history when expressive mode is offPossibly taken @ShivamSharma43 claimed this 23 days ago. Open
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
livekit/agents#7647 · 1 comment ·
Maintainers usually reply within 1 day
-
docs: openai.LLM vs openai.responses.LLM for HTTP OpenAI-compat hostsMay be free again @AbarnaaSree claimed this 62 days ago, and no pull request is open. Open
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
livekit/agents#7022 · 1 comment ·
Maintainers usually reply within 1 day
-
DataStreamAudioReceiver retains registrations after close and shares clear-buffer routing across roomsPossibly taken @Zheng-Changfu claimed this today. Open
Difficulty 4/5 3-5 days Newbie friendliness 20/100
Maintainers usually reply within 1 day
Similar issues
-
getchaintips leaves out the active tip once a header is staged on it, where Core always lists itOpen
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
btclib-org/btclib-node#2158 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 85/100
dbt-msft/dbt-sqlserver#897 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 65/100
Gathin23/cleanyourtv#2770 · 2 comments ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
vllm-project/vllm#64845 ·
Maintainers usually reply within 1 day
-
bug
Difficulty 2/5 1-3 hours Newbie friendliness 84/100
dekobon/big-code-analysis#1729 ·
Maintainers usually reply within 1 day