deepgeek/gradio-webrtc
2
0
Fork 0
mirror of https://github.com/HumanAIGC-Engineering/gradio-webrtc.git synced 2026-02-05 18:09:23 +08:00
Code Issues Packages Projects Releases Wiki Activity

ReplyOnPause and ReplyOnStopWords can be interrupted (#119)

Browse Source 
* Add all this code

* add code

* Fix demo

---------

Co-authored-by: Freddy Boulton <freddyboulton@hf-freddy.local>
This commit is contained in:
Freddy Boulton
2025-03-03 21:47:16 -05:00
committed by GitHub
parent 87954a62aa
commit 6ea54777af
13 changed files with 155 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
docs/userguide/audio.md
View File

@@ -3,6 +3,8 @@
Typically, you want to run a python function whenever a user has stopped speaking. This can be done by wrapping a python generator with the `ReplyOnPause` class and passing it to the `handler` argument of the `Stream` object. The `ReplyOnPause` class will handle the voice detection and turn taking logic automatically!
By default, the `ReplyOnPause` handler will allow you to interrupt the response at any time by speaking again. If you do not want to allow interruption, you can set the `can_interrupt` parameter to `False`.
=== "Code"
```python
from fastrtc import ReplyOnPause, Stream
@@ -33,13 +35,14 @@ Typically, you want to run a python function whenever a user has stopped speakin
You can also use an async generator with `ReplyOnPause`.
!!! tip "Parameters"
You can customize the voice detection parameters by passing in `algo_options` and `model_options` to the `ReplyOnPause` class.
You can customize the voice detection parameters by passing in `algo_options` and `model_options` to the `ReplyOnPause` class. Also, you can set the `can_interrupt` parameter to `False` to prevent the user from interrupting the response. By default, `can_interrupt` is `True`.
```python
from fastrtc import AlgoOptions, SileroVadOptions
stream = Stream(
handler=ReplyOnPause(
response,
can_interrupt=True,
algo_options=AlgoOptions(
audio_chunk_duration=0.6,
started_talking_threshold=0.2,