apunkt/dograh
1
0
Fork 0
mirror of https://github.com/dograh-hq/dograh.git synced 2026-07-04 10:52:17 +02:00
Code Issues Projects Releases Packages Wiki Activity

docs: fix dead entry points, add first-agent tutorial, explain unexplained features (#489)

Browse source 
* docs: fix dead entry points, add first-agent tutorial, explain gathered_context, VAD, E.164, Langfuse, Pipecat

* docs: fix broken anchors, correct HTTPS/VAD wording per review

* docs: tighten HTTPS wording and soften single-node claims per bot review
This commit is contained in:
Rushil 2026-07-02 12:47:48 +05:30 • committed by GitHub
parent dc98298b66
commit 97803b8121
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
18 changed files with 253 additions and 25 deletions
Download patch file Download diff file Expand all files Collapse all files

12
docs/configurations/interruption.mdx
View file

@ -9,6 +9,18 @@ Interruption handling controls whether the user can "barge in" and interrupt the
![Allow Interruption Toggle](../images/allow-interruption.png)
## What is VAD?
VAD (Voice Activity Detection) is what makes barge-in possible — it's the model that listens to the incoming audio stream and decides whether the caller is currently speaking. Dograh's pipeline runs [Silero VAD](https://github.com/snakers4/silero-vad) by default to detect the start and end of user speech in real time. Some realtime providers (such as OpenAI Realtime, Azure Realtime, and Grok Voice Agent) supply their own VAD signals, so local Silero VAD is skipped for those calls.
When **Allow Interruption** is enabled on a node, VAD is what triggers the interrupt: the moment it detects the caller has started talking, the bot's speech is cut off and the pipeline starts processing the new input. When interruption is disabled, VAD output for the user's mic is ignored until the bot finishes speaking.
VAD also feeds turn-taking more generally — it's part of how Dograh decides a caller has finished a turn and the agent should respond, independent of whether interruption is enabled.
<Note>
**Valid sample rates**: Dograh's pipeline runs VAD at either **8000 Hz** or **16000 Hz** — no other value is accepted. The overall pipeline sample rate is capped at 16 kHz to satisfy this. This only matters if you're integrating a [custom telephony provider](/integrations/telephony/custom#audio-format-considerations); telephony providers built into Dograh (Twilio, Vonage, Plivo, etc.) already negotiate a supported rate for you.
</Note>
## How It Works
Each node in your workflow has an **Allow Interruption** toggle:

2
docs/configurations/tracing.mdx
View file

@ -108,6 +108,8 @@ Every time an LLM call is made, the **entire conversation history up to that poi
## Setting Up Langfuse Tracing
[Langfuse](https://langfuse.com/docs) is an open-source LLM observability platform — it stores and visualizes traces (prompts, responses, tool calls) so you can debug and iterate on LLM-powered applications outside of Dograh's own trace viewer.
We provide seamless integration with Langfuse for tracing if you want to use your own account. This enables you to use the [playground feature of Langfuse](https://langfuse.com/docs/prompt-management/features/playground). This works on both managed and self-hosted Dograh deployments.
**Setup steps:**