Skip to content

fix: reject incomplete streaming transcriptions #289

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

Open
ch-z-hc wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix: reject incomplete streaming transcriptions #289

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
6 changes: 5 additions & 1 deletion src/commands/speech/transcribe.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -174,7 +174,11 @@ export default defineCommand({
}

if (!finished) {
process.stderr.write('[warning] Stream ended before the final event; the transcript may be incomplete.\n');
if (toStdout) process.stdout.write('\n');
throw new CLIError(
'Stream ended before the final event; the transcript may be incomplete.',
ExitCode.GENERAL,
);
}

// Assembled by index, per the API's contract — arrival order only shows
Expand Down
29 changes: 19 additions & 10 deletions test/commands/speech/transcribe.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -420,11 +420,18 @@ describe('speech transcribe command', () => {
}
});

it('warns on stderr when the stream ends without the final event', async () => {
it.each([
{ format: 'text' as const, ending: 'eof' },
{ format: 'text' as const, ending: 'done' },
{ format: 'json' as const, ending: 'eof' },
{ format: 'json' as const, ending: 'done' },
])('rejects incomplete $format transcription at $ending', async ({ format, ending }) => {
const { dir, filePath } = makeTempAudio();
const response = sseResponse([
{ data: '{"index":0,"delta":"truncated","finish":false}' },
]);
const response = new Response(
'data: {"index":0,"delta":"truncated","finish":false}\n\n'
+ (ending === 'done' ? 'data: [DONE]\n\n' : ''),
{ headers: { 'Content-Type': 'text/event-stream' } },
);

try {
await withStubbedFetch(
Expand All @@ -433,16 +440,18 @@ describe('speech transcribe command', () => {
let captured = '';
const stderr = await captureStderr(async () => {
captured = await captureStdout(async () => {
await transcribeCommand.execute(baseConfig, {
...baseFlags,
file: filePath,
stream: true,
await expect(transcribeCommand.execute({ ...baseConfig, output: format }, {
...baseFlags, file: filePath, stream: true,
})).rejects.toMatchObject({
name: 'CLIError',
exitCode: 1,
message: 'Stream ended before the final event; the transcript may be incomplete.',
});
});
});

expect(captured).toBe('truncated\n');
expect(stderr).toContain('Stream ended before the final event');
expect(captured).toBe(format === 'text' ? 'truncated\n' : '');
expect(stderr).not.toContain('[Duration:');
},
);
} finally {
Expand Down