Skip to content

Add language detect constraint #1212

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
yujonglee merged 6 commits into from
Jul 25, 2025
Merged

Add language detect constraint #1212

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
0462c39
wip
yujonglee Jul 25, 2025
69b8918
rust side done
yujonglee Jul 25, 2025
177eca9
fix
yujonglee Jul 25, 2025
2766aa8
one
yujonglee Jul 25, 2025
e369826
set translate
yujonglee Jul 25, 2025
9fdaa1a
i18n
yujonglee Jul 25, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 19 additions & 0 deletions Cargo.lock
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

3 changes: 2 additions & 1 deletion Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -149,7 +149,8 @@ inventory = "0.3.20"
serial_test = "3"
testcontainers-modules = "0.11.5"

axum = "0.8.1"
axum = "0.8.4"
axum-extra = "0.10.1"
tower = "0.5.2"
tower-http = "0.6.2"

Expand Down
5 changes: 4 additions & 1 deletion apps/app/server/src/native/listen/realtime.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,7 +29,10 @@ async fn websocket(socket: WebSocket, state: STTState, params: ListenParams) {

let (mut ws_sender, ws_receiver) = socket.split();

let mut stt = state.realtime_stt.for_language(params.language).await;
let mut stt = state
.realtime_stt
.for_language(params.languages.first().unwrap().clone())
.await;

let input_stream =
futures_util::stream::try_unfold(ws_receiver, |mut ws_receiver| async move {
Expand Down
5 changes: 4 additions & 1 deletion apps/app/server/src/native/listen/recorded.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,10 @@ pub async fn handler(
Query(params): Query<ListenParams>,
State(state): State<STTState>,
) -> impl IntoResponse {
let stt = state.recorded_stt.for_language(params.language).await;
let stt = state
.recorded_stt
.for_language(params.languages.first().unwrap().clone())
.await;

let input = RecordedSpeech::File("TODO".into());
let result = stt.transcribe(input).await.unwrap();
Expand Down
125 changes: 104 additions & 21 deletions apps/desktop/src/components/settings/views/general.tsx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,13 +2,17 @@ import { zodResolver } from "@hookform/resolvers/zod";
import { LANGUAGES_ISO_639_1 } from "@huggingface/languages";
import { Trans, useLingui } from "@lingui/react/macro";
import { useMutation, useQuery, useQueryClient } from "@tanstack/react-query";
import { Plus, X } from "lucide-react";
import { useEffect } from "react";
import { useForm } from "react-hook-form";
import { z } from "zod";

import { showModelSelectToast } from "@/components/toast/model-select";
import { commands } from "@/types";
import { commands as dbCommands, type ConfigGeneral } from "@hypr/plugin-db";
import { Badge } from "@hypr/ui/components/ui/badge";
import { Button } from "@hypr/ui/components/ui/button";
import { Command, CommandEmpty, CommandGroup, CommandInput, CommandItem } from "@hypr/ui/components/ui/command";
import {
Form,
FormControl,
Expand All @@ -19,7 +23,8 @@ import {
FormMessage,
} from "@hypr/ui/components/ui/form";
import { Input } from "@hypr/ui/components/ui/input";
import { Select, SelectContent, SelectItem, SelectTrigger, SelectValue } from "@hypr/ui/components/ui/select";
import { Popover, PopoverContent, PopoverTrigger } from "@hypr/ui/components/ui/popover";
import { Select, SelectTrigger, SelectValue } from "@hypr/ui/components/ui/select";
import { Switch } from "@hypr/ui/components/ui/switch";

type ISO_639_1_CODE = keyof typeof LANGUAGES_ISO_639_1;
Expand Down Expand Up @@ -71,6 +76,7 @@ const SUPPORTED_LANGUAGES: ISO_639_1_CODE[] = [
const schema = z.object({
autostart: z.boolean().optional(),
displayLanguage: z.enum(SUPPORTED_LANGUAGES as [string, ...string[]]),
spokenLanguages: z.array(z.enum(SUPPORTED_LANGUAGES as [string, ...string[]])).min(1),
telemetryConsent: z.boolean().optional(),
jargons: z.string(),
saveRecordings: z.boolean().optional(),
Expand All @@ -95,6 +101,7 @@ export default function General() {
defaultValues: {
autostart: false,
displayLanguage: "en",
spokenLanguages: ["en"],
telemetryConsent: true,
jargons: "",
saveRecordings: true,
Expand All @@ -106,6 +113,7 @@ export default function General() {
form.reset({
autostart: config.data.general.autostart ?? false,
displayLanguage: config.data.general.display_language ?? "en",
spokenLanguages: config.data.general.spoken_languages ?? ["en"],
telemetryConsent: config.data.general.telemetry_consent ?? true,
jargons: (config.data.general.jargons ?? []).join(", "),
saveRecordings: config.data.general.save_recordings ?? true,
Expand All @@ -123,6 +131,7 @@ export default function General() {
const nextGeneral: ConfigGeneral = {
autostart: v.autostart ?? false,
display_language: v.displayLanguage,
spoken_languages: v.spokenLanguages,
telemetry_consent: v.telemetryConsent ?? true,
jargons: v.jargons.split(",").map((jargon) => jargon.trim()).filter(Boolean),
save_recordings: v.saveRecordings ?? true,
Expand Down Expand Up @@ -163,7 +172,7 @@ export default function General() {
return (
<div>
<Form {...form}>
<form className="space-y-6">
<form className="space-y-8">
<FormField
control={form.control}
name="saveRecordings"
Expand Down Expand Up @@ -224,31 +233,105 @@ export default function General() {
<FormItem className="flex flex-row items-center justify-between">
<div className="space-y-0.5">
<FormLabel>
<Trans>Language</Trans>
<Trans>Display language</Trans>
</FormLabel>
<FormDescription>
<Trans>Choose your preferred language of use</Trans>
<Trans>Primary language for the interface</Trans>
</FormDescription>
</div>
<FormControl>
<Select
onValueChange={field.onChange}
value={field.value}
disabled
value="en"
>
<SelectTrigger className="w-[200px]">
<SelectValue placeholder="Select language" />
<SelectValue>
{LANGUAGES_ISO_639_1["en"].name}
</SelectValue>
</SelectTrigger>
<SelectContent className="max-h-64">
{SUPPORTED_LANGUAGES.map((lang) => (
<SelectItem key={lang} value={lang}>
{LANGUAGES_ISO_639_1[lang].name}
</SelectItem>
))}
</SelectContent>
</Select>
</FormControl>
{/* FormMessage is usually displayed below the control, might need separate handling if it must be in-row and an error occurs */}
{/* <FormMessage /> */}
</FormItem>
)}
/>

<FormField
control={form.control}
name="spokenLanguages"
render={({ field }) => (
<FormItem>
<div className="space-y-0.5">
<FormLabel>
<Trans>Spoken languages</Trans>
</FormLabel>
<FormDescription>
<Trans>Select languages you speak for better transcription</Trans>
</FormDescription>
</div>
<FormControl>
<div className="flex items-center gap-2">
<div className="flex-1 flex flex-wrap gap-2 min-h-[38px] p-2 border rounded-md">
{field.value.map((langCode) => (
<Badge
key={langCode}
variant="secondary"
className="flex items-center gap-1 px-2 py-0.5 text-xs bg-muted"
>
{LANGUAGES_ISO_639_1[langCode as ISO_639_1_CODE]?.name || langCode}
<Button
type="button"
variant="ghost"
size="sm"
className="h-3 w-3 p-0 hover:bg-transparent ml-0.5"
onClick={() => {
const newLanguages = field.value.filter((lang) => lang !== langCode);
field.onChange(newLanguages);
mutation.mutate(form.getValues());
}}
>
<X className="h-2.5 w-2.5" />
</Button>
</Badge>
))}
</div>
<Popover>
<PopoverTrigger asChild>
<Button
type="button"
variant="outline"
size="icon"
className="h-[38px] w-[38px]"
>
<Plus className="h-4 w-4" />
</Button>
</PopoverTrigger>
<PopoverContent className="w-[220px] p-0" align="end">
<Command>
<CommandInput placeholder="Search languages..." className="h-9" />
<CommandEmpty>No language found.</CommandEmpty>
<CommandGroup className="max-h-[200px] overflow-auto">
{SUPPORTED_LANGUAGES.filter(
(lang) => !field.value.includes(lang),
).map((lang) => (
<CommandItem
key={lang}
onSelect={() => {
if (!field.value.includes(lang)) {
field.onChange([...field.value, lang]);
mutation.mutate(form.getValues());
}
}}
>
{LANGUAGES_ISO_639_1[lang].name}
</CommandItem>
))}
</CommandGroup>
</Command>
</PopoverContent>
</Popover>
</div>
</FormControl>
<FormMessage />
</FormItem>
)}
/>
Expand All @@ -258,13 +341,13 @@ export default function General() {
name="jargons"
render={({ field }) => (
<FormItem>
<div>
<div className="space-y-0.5">
<FormLabel>
<Trans>Jargons</Trans>
<Trans>Custom Vocabulary</Trans>
</FormLabel>
<FormDescription>
<Trans>
You can make Hyprnote takes these words into account when transcribing
Add specific terms or jargon for improved transcription accuracy
</Trans>
</FormDescription>
</div>
Expand All @@ -273,9 +356,9 @@ export default function General() {
{...field}
onBlur={() => mutation.mutate(form.getValues())}
placeholder={t({
id: "Type jargons (e.g., Blitz Meeting, PaC Squad)",
id: "Type terms separated by commas (e.g., Blitz Meeting, PaC Squad)",
})}
className="focus-visible:ring-0 focus-visible:ring-offset-0"
className="focus-visible:ring-1 focus-visible:ring-ring"
/>
</FormControl>
<FormMessage />
Expand Down
Loading
Loading