selfhostyourtech/reflector
2
0
Fork 0
mirror of https://github.com/Monadical-SAS/reflector.git synced 2025-12-20 12:19:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat: add transcript format parameter to GET endpoint (#709)

Browse Source 
* feat: add transcript format parameter to GET endpoint

Add transcript_format query parameter to /v1/transcripts/{id} endpoint
with support for multiple output formats using discriminated unions.

Formats supported:
- text: Plain speaker dialogue (default)
- text-timestamped: Dialogue with [MM:SS] timestamps
- webvtt-named: WebVTT subtitles with participant names
- json: Structured segments with full metadata

Response models use Pydantic discriminated unions with transcript_format
as discriminator field. POST/PATCH endpoints return GetTranscriptWithParticipants
for minimal responses. GET endpoint returns format-specific models.

* Copy transcript format

* Regenerate types

* Fix transcript formats

* Don't throw inside try

* Remove any type

* Toast share copy errors

* transcript_format exhaustiveness and python idiomatic assert_never

* format_timestamp_mmss clear type definition

* Rename seconds_to_timestamp

* Test transcript format with overlapping speakers

* exact match for vtt multispeaker test

---------

Co-authored-by: Sergey Mankovsky <sergey@monadical.com>
Co-authored-by: Igor Loskutov <igor.loskutoff@gmail.com>
This commit is contained in:
Mathieu Virbel
2025-11-26 11:51:14 -06:00
committed by GitHub
parent 3aef926203
commit f6ca07505f
12 changed files with 1625 additions and 138 deletions
Download Patch File Download Diff File Expand all files Collapse all files

130
www/app/(app)/transcripts/shareCopy.tsx
View File

@@ -1,14 +1,16 @@
import { useState } from "react";
import type { components } from "../../reflector-api";
type GetTranscript = components["schemas"]["GetTranscript"];
import type { components, operations } from "../../reflector-api";
type GetTranscriptWithParticipants =
components["schemas"]["GetTranscriptWithParticipants"];
type GetTranscriptTopic = components["schemas"]["GetTranscriptTopic"];
import { Button, BoxProps, Box } from "@chakra-ui/react";
import { buildTranscriptWithTopics } from "./buildTranscriptWithTopics";
import { useTranscriptParticipants } from "../../lib/apiHooks";
import { Button, BoxProps, Box, Menu, Text } from "@chakra-ui/react";
import { LuChevronDown } from "react-icons/lu";
import { client } from "../../lib/apiClient";
import { toaster } from "../../components/ui/toaster";
type ShareCopyProps = {
finalSummaryElement: HTMLDivElement | null;
transcript: GetTranscript;
transcript: GetTranscriptWithParticipants;
topics: GetTranscriptTopic[];
};
@@ -20,11 +22,33 @@ export default function ShareCopy({
}: ShareCopyProps & BoxProps) {
const [isCopiedSummary, setIsCopiedSummary] = useState(false);
const [isCopiedTranscript, setIsCopiedTranscript] = useState(false);
const participantsQuery = useTranscriptParticipants(transcript?.id || null);
const [isCopying, setIsCopying] = useState(false);
type ApiTranscriptFormat = NonNullable<
operations["v1_transcript_get"]["parameters"]["query"]
>["transcript_format"];
const TRANSCRIPT_FORMATS = [
"text",
"text-timestamped",
"webvtt-named",
"json",
] as const satisfies ApiTranscriptFormat[];
type TranscriptFormat = (typeof TRANSCRIPT_FORMATS)[number];
const TRANSCRIPT_FORMAT_LABELS: { [k in TranscriptFormat]: string } = {
text: "Plain text",
"text-timestamped": "Text + timestamps",
"webvtt-named": "WebVTT (named)",
json: "JSON",
};
const formatOptions = TRANSCRIPT_FORMATS.map((f) => ({
value: f,
label: TRANSCRIPT_FORMAT_LABELS[f],
}));
const onCopySummaryClick = () => {
const text_to_copy = finalSummaryElement?.innerText;
if (text_to_copy) {
navigator.clipboard.writeText(text_to_copy).then(() => {
setIsCopiedSummary(true);
@@ -34,27 +58,91 @@ export default function ShareCopy({
}
};
const onCopyTranscriptClick = () => {
const text_to_copy =
buildTranscriptWithTopics(
topics || [],
participantsQuery?.data || null,
transcript?.title || null,
) || "";
const onCopyTranscriptFormatClick = async (format: TranscriptFormat) => {
try {
setIsCopying(true);
const { data, error } = await client.GET(
"/v1/transcripts/{transcript_id}",
{
params: {
path: { transcript_id: transcript.id },
query: { transcript_format: format },
},
},
);
if (error) {
console.error("Failed to copy transcript:", error);
toaster.create({
duration: 3000,
render: () => (
<Box bg="red.500" color="white" px={4} py={3} borderRadius="md">
<Text fontWeight="bold">Error</Text>
<Text fontSize="sm">Failed to fetch transcript</Text>
</Box>
),
});
return;
}
text_to_copy &&
navigator.clipboard.writeText(text_to_copy).then(() => {
const copiedText =
format === "json"
? JSON.stringify(data?.transcript ?? {}, null, 2)
: String(data?.transcript ?? "");
if (copiedText) {
await navigator.clipboard.writeText(copiedText);
setIsCopiedTranscript(true);
// Reset the copied state after 2 seconds
setTimeout(() => setIsCopiedTranscript(false), 2000);
}
} catch (e) {
console.error("Failed to copy transcript:", e);
toaster.create({
duration: 3000,
render: () => (
<Box bg="red.500" color="white" px={4} py={3} borderRadius="md">
<Text fontWeight="bold">Error</Text>
<Text fontSize="sm">Failed to copy transcript</Text>
</Box>
),
});
} finally {
setIsCopying(false);
}
};
return (
<Box {...boxProps}>
<Button onClick={onCopyTranscriptClick} mr={2} variant="subtle">
{isCopiedTranscript ? "Copied!" : "Copy Transcript"}
</Button>
<Menu.Root
closeOnSelect={true}
lazyMount={true}
positioning={{ gutter: 4 }}
>
<Menu.Trigger asChild>
<Button
mr={2}
variant="subtle"
loading={isCopying}
loadingText="Copying..."
>
{isCopiedTranscript ? "Copied!" : "Copy Transcript"}
<LuChevronDown style={{ marginLeft: 6 }} />
</Button>
</Menu.Trigger>
<Menu.Positioner>
<Menu.Content>
{formatOptions.map((opt) => (
<Menu.Item
key={opt.value}
value={opt.value}
_hover={{ backgroundColor: "gray.100" }}
onClick={() => onCopyTranscriptFormatClick(opt.value)}
>
{opt.label}
</Menu.Item>
))}
</Menu.Content>
</Menu.Positioner>
</Menu.Root>
<Button onClick={onCopySummaryClick} variant="subtle">
{isCopiedSummary ? "Copied!" : "Copy Summary"}
</Button>

5
www/app/lib/authBackend.ts
View File

@@ -32,6 +32,11 @@ async function getUserId(accessToken: string): Promise<string | null> {
});
if (!response.ok) {
try {
console.error(await response.text());
} catch (e) {
console.error("Failed to parse error response", e);
}
return null;
}

486
www/app/reflector-api.d.ts vendored
View File

@@ -696,7 +696,7 @@ export interface paths {
patch?: never;
trace?: never;
};
"/v1/webhook": {
"/v1/daily/webhook": {
parameters: {
query?: never;
header?: never;
@@ -708,6 +708,27 @@ export interface paths {
/**
* Webhook
* @description Handle Daily webhook events.
*
* Example webhook payload:
* {
* "version": "1.0.0",
* "type": "recording.ready-to-download",
* "id": "rec-rtd-c3df927c-f738-4471-a2b7-066fa7e95a6b-1692124192",
* "payload": {
* "recording_id": "08fa0b24-9220-44c5-846c-3f116cf8e738",
* "room_name": "Xcm97xRZ08b2dePKb78g",
* "start_ts": 1692124183,
* "status": "finished",
* "max_participants": 1,
* "duration": 9,
* "share_token": "ntDCL5k98Ulq", #gitleaks:allow
* "s3_key": "api-test-1j8fizhzd30c/Xcm97xRZ08b2dePKb78g/1692124183028"
* },
* "event_ts": 1692124192
* }
*
* Daily.co circuit-breaker: After 3+ failed responses (4xx/5xx), webhook
* state→FAILED, stops sending events. Reset: scripts/recreate_daily_webhook.py
*/
post: operations["v1_webhook"];
delete?: never;
@@ -899,81 +920,11 @@ export interface components {
target_language: string;
source_kind?: components["schemas"]["SourceKind"] | null;
};
/**
* DailyWebhookEvent
* @description Daily webhook event structure.
*/
DailyWebhookEvent: {
/** Type */
type: string;
/** Id */
id: string;
/** Ts */
ts: number;
/** Data */
data: {
[key: string]: unknown;
};
};
/** DeletionStatus */
DeletionStatus: {
/** Status */
status: string;
};
/** GetTranscript */
GetTranscript: {
/** Id */
id: string;
/** User Id */
user_id: string | null;
/** Name */
name: string;
/**
* Status
* @enum {string}
*/
status:
| "idle"
| "uploaded"
| "recording"
| "processing"
| "error"
| "ended";
/** Locked */
locked: boolean;
/** Duration */
duration: number;
/** Title */
title: string | null;
/** Short Summary */
short_summary: string | null;
/** Long Summary */
long_summary: string | null;
/** Created At */
created_at: string;
/**
* Share Mode
* @default private
*/
share_mode: string;
/** Source Language */
source_language: string | null;
/** Target Language */
target_language: string | null;
/** Reviewed */
reviewed: boolean;
/** Meeting Id */
meeting_id: string | null;
source_kind: components["schemas"]["SourceKind"];
/** Room Id */
room_id?: string | null;
/** Room Name */
room_name?: string | null;
/** Audio Deleted */
audio_deleted?: boolean | null;
/** Participants */
participants: components["schemas"]["TranscriptParticipant"][] | null;
};
/** GetTranscriptMinimal */
GetTranscriptMinimal: {
/** Id */
@@ -1105,6 +1056,345 @@ export interface components {
*/
words_per_speaker: components["schemas"]["SpeakerWords"][];
};
/**
* GetTranscriptWithJSON
* @description Transcript response as structured JSON segments.
*
* Format: Array of segment objects with speaker info, text, and timing.
* Example:
* [
* {
* "speaker": 0,
* "speaker_name": "John Smith",
* "text": "Hello everyone",
* "start": 0.0,
* "end": 5.0
* }
* ]
*/
GetTranscriptWithJSON: {
/** Id */
id: string;
/** User Id */
user_id: string | null;
/** Name */
name: string;
/**
* Status
* @enum {string}
*/
status:
| "idle"
| "uploaded"
| "recording"
| "processing"
| "error"
| "ended";
/** Locked */
locked: boolean;
/** Duration */
duration: number;
/** Title */
title: string | null;
/** Short Summary */
short_summary: string | null;
/** Long Summary */
long_summary: string | null;
/** Created At */
created_at: string;
/**
* Share Mode
* @default private
*/
share_mode: string;
/** Source Language */
source_language: string | null;
/** Target Language */
target_language: string | null;
/** Reviewed */
reviewed: boolean;
/** Meeting Id */
meeting_id: string | null;
source_kind: components["schemas"]["SourceKind"];
/** Room Id */
room_id?: string | null;
/** Room Name */
room_name?: string | null;
/** Audio Deleted */
audio_deleted?: boolean | null;
/** Participants */
participants: components["schemas"]["TranscriptParticipant"][] | null;
/**
* @description discriminator enum property added by openapi-typescript
* @enum {string}
*/
transcript_format: "json";
/** Transcript */
transcript: components["schemas"]["TranscriptSegment"][];
};
/** GetTranscriptWithParticipants */
GetTranscriptWithParticipants: {
/** Id */
id: string;
/** User Id */
user_id: string | null;
/** Name */
name: string;
/**
* Status
* @enum {string}
*/
status:
| "idle"
| "uploaded"
| "recording"
| "processing"
| "error"
| "ended";
/** Locked */
locked: boolean;
/** Duration */
duration: number;
/** Title */
title: string | null;
/** Short Summary */
short_summary: string | null;
/** Long Summary */
long_summary: string | null;
/** Created At */
created_at: string;
/**
* Share Mode
* @default private
*/
share_mode: string;
/** Source Language */
source_language: string | null;
/** Target Language */
target_language: string | null;
/** Reviewed */
reviewed: boolean;
/** Meeting Id */
meeting_id: string | null;
source_kind: components["schemas"]["SourceKind"];
/** Room Id */
room_id?: string | null;
/** Room Name */
room_name?: string | null;
/** Audio Deleted */
audio_deleted?: boolean | null;
/** Participants */
participants: components["schemas"]["TranscriptParticipant"][] | null;
};
/**
* GetTranscriptWithText
* @description Transcript response with plain text format.
*
* Format: Speaker names followed by their dialogue, one line per segment.
* Example:
* John Smith: Hello everyone
* Jane Doe: Hi there
*/
GetTranscriptWithText: {
/** Id */
id: string;
/** User Id */
user_id: string | null;
/** Name */
name: string;
/**
* Status
* @enum {string}
*/
status:
| "idle"
| "uploaded"
| "recording"
| "processing"
| "error"
| "ended";
/** Locked */
locked: boolean;
/** Duration */
duration: number;
/** Title */
title: string | null;
/** Short Summary */
short_summary: string | null;
/** Long Summary */
long_summary: string | null;
/** Created At */
created_at: string;
/**
* Share Mode
* @default private
*/
share_mode: string;
/** Source Language */
source_language: string | null;
/** Target Language */
target_language: string | null;
/** Reviewed */
reviewed: boolean;
/** Meeting Id */
meeting_id: string | null;
source_kind: components["schemas"]["SourceKind"];
/** Room Id */
room_id?: string | null;
/** Room Name */
room_name?: string | null;
/** Audio Deleted */
audio_deleted?: boolean | null;
/** Participants */
participants: components["schemas"]["TranscriptParticipant"][] | null;
/**
* @description discriminator enum property added by openapi-typescript
* @enum {string}
*/
transcript_format: "text";
/** Transcript */
transcript: string;
};
/**
* GetTranscriptWithTextTimestamped
* @description Transcript response with timestamped text format.
*
* Format: [MM:SS] timestamp prefix before each speaker and dialogue.
* Example:
* [00:00] John Smith: Hello everyone
* [00:05] Jane Doe: Hi there
*/
GetTranscriptWithTextTimestamped: {
/** Id */
id: string;
/** User Id */
user_id: string | null;
/** Name */
name: string;
/**
* Status
* @enum {string}
*/
status:
| "idle"
| "uploaded"
| "recording"
| "processing"
| "error"
| "ended";
/** Locked */
locked: boolean;
/** Duration */
duration: number;
/** Title */
title: string | null;
/** Short Summary */
short_summary: string | null;
/** Long Summary */
long_summary: string | null;
/** Created At */
created_at: string;
/**
* Share Mode
* @default private
*/
share_mode: string;
/** Source Language */
source_language: string | null;
/** Target Language */
target_language: string | null;
/** Reviewed */
reviewed: boolean;
/** Meeting Id */
meeting_id: string | null;
source_kind: components["schemas"]["SourceKind"];
/** Room Id */
room_id?: string | null;
/** Room Name */
room_name?: string | null;
/** Audio Deleted */
audio_deleted?: boolean | null;
/** Participants */
participants: components["schemas"]["TranscriptParticipant"][] | null;
/**
* @description discriminator enum property added by openapi-typescript
* @enum {string}
*/
transcript_format: "text-timestamped";
/** Transcript */
transcript: string;
};
/**
* GetTranscriptWithWebVTTNamed
* @description Transcript response in WebVTT subtitle format with participant names.
*
* Format: Standard WebVTT with voice tags using participant names.
* Example:
* WEBVTT
*
* 00:00:00.000 --> 00:00:05.000
* <v John Smith>Hello everyone
*/
GetTranscriptWithWebVTTNamed: {
/** Id */
id: string;
/** User Id */
user_id: string | null;
/** Name */
name: string;
/**
* Status
* @enum {string}
*/
status:
| "idle"
| "uploaded"
| "recording"
| "processing"
| "error"
| "ended";
/** Locked */
locked: boolean;
/** Duration */
duration: number;
/** Title */
title: string | null;
/** Short Summary */
short_summary: string | null;
/** Long Summary */
long_summary: string | null;
/** Created At */
created_at: string;
/**
* Share Mode
* @default private
*/
share_mode: string;
/** Source Language */
source_language: string | null;
/** Target Language */
target_language: string | null;
/** Reviewed */
reviewed: boolean;
/** Meeting Id */
meeting_id: string | null;
source_kind: components["schemas"]["SourceKind"];
/** Room Id */
room_id?: string | null;
/** Room Name */
room_name?: string | null;
/** Audio Deleted */
audio_deleted?: boolean | null;
/** Participants */
participants: components["schemas"]["TranscriptParticipant"][] | null;
/**
* @description discriminator enum property added by openapi-typescript
* @enum {string}
*/
transcript_format: "webvtt-named";
/** Transcript */
transcript: string;
};
/** HTTPValidationError */
HTTPValidationError: {
/** Detail */
@@ -1233,7 +1523,6 @@ export interface components {
} | null;
/**
* Platform
* @default whereby
* @enum {string}
*/
platform: "whereby" | "daily";
@@ -1325,7 +1614,6 @@ export interface components {
ics_last_etag?: string | null;
/**
* Platform
* @default whereby
* @enum {string}
*/
platform: "whereby" | "daily";
@@ -1377,7 +1665,6 @@ export interface components {
ics_last_etag?: string | null;
/**
* Platform
* @default whereby
* @enum {string}
*/
platform: "whereby" | "daily";
@@ -1523,6 +1810,24 @@ export interface components {
speaker: number | null;
/** Name */
name: string;
/** User Id */
user_id?: string | null;
};
/**
* TranscriptSegment
* @description A single transcript segment with speaker and timing information.
*/
TranscriptSegment: {
/** Speaker */
speaker: number;
/** Speaker Name */
speaker_name: string;
/** Text */
text: string;
/** Start */
start: number;
/** End */
end: number;
};
/** UpdateParticipant */
UpdateParticipant: {
@@ -2311,7 +2616,7 @@ export interface operations {
[name: string]: unknown;
};
content: {
"application/json": components["schemas"]["GetTranscript"];
"application/json": components["schemas"]["GetTranscriptWithParticipants"];
};
};
/** @description Validation Error */
@@ -2369,7 +2674,13 @@ export interface operations {
};
v1_transcript_get: {
parameters: {
query?: never;
query?: {
transcript_format?:
| "text"
| "text-timestamped"
| "webvtt-named"
| "json";
};
header?: never;
path: {
transcript_id: string;
@@ -2384,7 +2695,11 @@ export interface operations {
[name: string]: unknown;
};
content: {
"application/json": components["schemas"]["GetTranscript"];
"application/json":
| components["schemas"]["GetTranscriptWithText"]
| components["schemas"]["GetTranscriptWithTextTimestamped"]
| components["schemas"]["GetTranscriptWithWebVTTNamed"]
| components["schemas"]["GetTranscriptWithJSON"];
};
};
/** @description Validation Error */
@@ -2450,7 +2765,7 @@ export interface operations {
[name: string]: unknown;
};
content: {
"application/json": components["schemas"]["GetTranscript"];
"application/json": components["schemas"]["GetTranscriptWithParticipants"];
};
};
/** @description Validation Error */
@@ -3256,11 +3571,7 @@ export interface operations {
path?: never;
cookie?: never;
};
requestBody: {
content: {
"application/json": components["schemas"]["DailyWebhookEvent"];
};
};
requestBody?: never;
responses: {
/** @description Successful Response */
200: {
@@ -3271,15 +3582,6 @@ export interface operations {
"application/json": unknown;
};
};
/** @description Validation Error */
422: {
headers: {
[name: string]: unknown;
};
content: {
"application/json": components["schemas"]["HTTPValidationError"];
};
};
};
};
}