llama.cpp
ui: voice messages and dictation with live waveforms and transcripts
#32531
Merged
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Overview
Commits
48
Changes
View On
GitHub
ui: voice messages and dictation with live waveforms and transcripts
#32531
allozaur
merged 48 commits into
ggml-org:master
from
allozaur:allozaur/feat/voice-mode-upstream
allozaur
force pushed
from
6fbc1c36
to
bd5ddd7b
2 days ago
github-actions
added
documentation
github-actions
added
devops
github-actions
added
server/ui
allozaur
force pushed
from
b86f6835
to
d5a945ce
2 days ago
allozaur
force pushed
from
2c6d00da
to
912f32c8
2 days ago
allozaur
force pushed
from
defe60da
to
b4516fab
1 day ago
allozaur
marked this pull request as ready for review
20 hours ago
allozaur
requested a review
20 hours ago
allozaur
requested a review
20 hours ago
allozaur
requested a review
from
ggerganov
19 hours ago
allozaur
requested a review
from
ServeurpersoCom
19 hours ago
allozaur
requested a review
from
ngxson
19 hours ago
CISC
approved these changes on 2026-11-20
allozaur
force pushed
from
ab02fb55
to
6fa78b56
19 hours ago
github-actions
added
video demo
allozaur
force pushed
from
6fa78b56
to
69a320eb
18 hours ago
ui: add a voice engine for capture, peaks and transcription
58369fb8
ui: add voice waveform and transcript components to Llama UI
91b3e58b
ui: add stories and tests for the voice components
0b6052da
ui: store voice messages with their transcript and send it to models …
e55bfecc
ui: add Voice settings and resolve what the microphone does per model
3ea381cd
ui: record voice messages and dictation from the chat form
7a5d8adb
ui: draw audio attachments as playable waveforms with their transcript
5fffee01
ui: drop the old audio recorder and proxy realtime transcription in dev
14c908f8
ui: dictate with a model that is already loaded when the chat model c…
d0b4c779
ui: fill the recording waveform, and open voice message transcripts
ff88a4db
ui: record voice without ducking other audio, from the first word
47a110f1
ui: show voice text live in the message box, and follow transcripts a…
cd8c0771
ui: let the capture test start with silence
ecbc98e7
ui: show live voice text sooner, and keep silence from becoming "yeah"
832b2148
ui: transcribe a recording cut at pauses once more as a whole
12c3e4f7
ui: give live text over HTTP for models without a realtime session
90661032
ui: say when the browser processes the microphone despite the raw req…
7392246d
ui: keep the microphone beside the stop and send buttons
d8b158bc
ui: show the voice components' stories with Llama UI
0a53262f
ui: add a noise suppression setting for voice input
5aaa447d
ui: keep live voice text coming when the room is never quiet
9c5ee835
ui: add echo cancellation and automatic gain settings for voice input
228b2256
ui: put the microphone after the stop button, filled like it
9f3c6a52
ui: say when dictated text is being refined, and send it refined
ee7b23ae
ui: keep the send button in place and put the microphone beside it
69fe9d93
ui: let a recording be sent while a reply streams, once it has words
163610b4
ui: raise quiet speech for transcription, and say when it is too quiet
bef4a07a
ui: keep the voice engine free of app code
dbc9809d
ui: start voice sessions from one voice service
b8cbca94
ui: tell speech from the room in one place
ef3f201c
ui: let any text field take dictation
3f8ef65f
ui: move the recording row into Llama UI
6011b740
ui: keep the voice engine's Svelte recorder behind its own entry
525fc5aa
ui: move voice input into its own package, Llama Voice
0ef6a25c
ui: show which model does what in voice input
4bfc0153
ui: warn of a quiet microphone inside the recording row
262cdf23
ui: open the microphone while the pointer is on its button
33887448
ui: tidy voice input after review
05e704f1
ui: pin Llama Voice's dev dependencies to Llama UI's versions
49144bf6
ui: let demos speak into a synthetic microphone
21395cd6
ui: keep the microphone for voice input, off the audio files entry
25fcc2aa
ui: set voice input on the slimmed design tokens
ae571b19
allozaur
force pushed
from
69a320eb
to
ae571b19
18 hours ago
allozaur
commented on 2026-11-20
allozaur
commented on 2026-11-20
ui: name voice states, platform strings, patterns and tunings
3927a837
ui: keep all of Llama UI's tests under tests
8a87b13a
ui: show a transcript's words without the model's own format
c6e0c1e2
ui: dictate when a speech-to-text model is picked for the conversation
d0fe3fa6
ui: say why a live-only model cannot transcribe behind an API key
9b198402
ngxson
approved these changes on 2026-11-20
ui: pick the dictation model with a microphone on its row
eebea0c8
ServeurpersoCom
approved these changes on 2026-11-20
allozaur
merged
79e81b0b
into master
13 hours ago
Login to write a write a comment.
Login via GitHub
Reviewers
ServeurpersoCom
ngxson
CISC
ggerganov
Assignees
No one assigned
Labels
documentation
devops
server/ui
video demo
Milestone
No milestone
Login to write a write a comment.
Login via GitHub