Download this bank as PDF

Listen first — sample clip types

Synthetic (TTS) examples — a guide to the shape of each clip type, not a voice to imitate.

Code-switch
Disfluency
Pause

German recording battery

Thank you for recording these. There are sentences below; each row tells you how to read it — plain ones naturally, and the marked code-switch / disfluency / pause / one-word clips following the note in their row. Listen to the three samples above first to hear what those clip types sound like.

How to record each clip

  1. Find a quiet room, start recording once (Voice Memos or equivalent is fine), and keep the phone recording the whole way through — you don’t stop and start between sentences, it’s all one recording. Read down the list in order.
  2. Leave a clear gap of about five seconds of silence between one sentence and the next — just wait quietly, then start the next one. Those gaps are how the single recording gets split back into individual clips afterwards, so make them clearly longer than any pause inside a sentence. If you fumble a sentence, just pause, say the whole sentence again, and carry on — the last clean take of each is what’s used.
  3. Read each sentence naturally, at your own pace. If you’re not fluent in the language, that’s fine — read it as a learner would: slow down, mispronounce things, self-correct. That IS the population this benchmark measures. Don’t over-produce it — hesitation and imperfection are the data, not noise to eliminate.
  4. For clips marked code-switch: say the [[bracketed]] word or phrase in English, out loud, right in the middle of the sentence — like you genuinely forgot or never learned the word in the target language. Don’t translate it.
  5. For the disfluency clips: follow their instruction (read through an inserted filler word naturally, cut yourself off at the em-dash and restart the word right after it, read slowly and deliberately, or pause partway through). The short pause inside one of these sentences is part of the sentence — keep it shorter than the five-second gap you leave between sentences.
  6. For the robustness probe clips: follow them exactly — either say only the single word and nothing else, or (for the silent one) stay silent for a few seconds. These exist to catch a model hallucinating words that were never said, so it matters that nothing extra gets said — then leave your normal between-sentence gap and move on to the next.

The battery — all 48 clips

# Type Say Instruction
1 code-switch (how-do-you-say) äh wie sagt man [[discount]] auf Deutsch ? Read the sentence, but say the [[bracketed]] word in English, like you’re asking how to say it.
2 code-switch (how-do-you-say) wie heißt [[schedule]] nochmal auf Deutsch ? Read the sentence, but say the [[bracketed]] word in English, like you’re asking how to say it.
3 code-switch (how-do-you-say) ich weiß nicht wie man [[appointment]] auf Deutsch sagt Read the sentence, but say the [[bracketed]] word in English, like you’re asking how to say it.
4 code-switch (how-do-you-say) äh wie sagt man [[deadline]] auf Deutsch ? Read the sentence, but say the [[bracketed]] word in English, like you’re asking how to say it.
5 code-switch (how-do-you-say) wie heißt [[day off]] nochmal auf Deutsch ? Read the sentence, but say the [[bracketed]] word in English, like you’re asking how to say it.
6 code-switch (how-do-you-say) ich weiß nicht wie man [[grocery store]] auf Deutsch sagt Read the sentence, but say the [[bracketed]] word in English, like you’re asking how to say it.
7 code-switch (noun) ich habe meinen [[insurance]] zu Hause vergessen Read the sentence, but say the [[bracketed]] word in English, like you forgot the German word for it.
8 code-switch (noun) ich brauche einen [[discount]] für das Gespräch morgen Read the sentence, but say the [[bracketed]] word in English, like you forgot the German word for it.
9 code-switch (noun) wir haben nächste Woche einen [[receipt]] glaube ich Read the sentence, but say the [[bracketed]] word in English, like you forgot the German word for it.
10 code-switch (noun) ich habe morgen früh einen äh [[schedule]] beim Arzt Read the sentence, but say the [[bracketed]] word in English, like you forgot the German word for it.
11 code-switch (noun) ich brauche einen [[laundry]] für das Gespräch morgen Read the sentence, but say the [[bracketed]] word in English, like you forgot the German word for it.
12 code-switch (noun) wir haben nächste Woche einen [[water bottle]] glaube ich Read the sentence, but say the [[bracketed]] word in English, like you forgot the German word for it.
13 code-switch (clause) ich wollte sagen dass äh [[actually I don’t know the word]] Read the sentence, but say the [[bracketed]] phrase in English, like you switched languages mid-thought.
14 code-switch (clause) das ist ein bisschen ähm [[how do I explain this]] schwer zu sagen Read the sentence, but say the [[bracketed]] phrase in English, like you switched languages mid-thought.
15 code-switch (clause) gestern war ich auf dem Markt und äh [[wait I forgot the word]] Read the sentence, but say the [[bracketed]] phrase in English, like you switched languages mid-thought.
16 code-switch (clause) sie hat mich gefragt ob äh [[sorry let me think for a second]] Read the sentence, but say the [[bracketed]] phrase in English, like you switched languages mid-thought.
17 robustness probe (one-word) genau. Say only this one word, then stop — nothing else before or after.
18 robustness probe (one-word) ja. Say only this one word, then stop — nothing else before or after.
19 disfluency (pause) Dieses Medikament kann Schwindel als Nebenwirkung verursachen. Read this sentence, but pause for about 5 seconds partway through, like you’re thinking mid-sentence, then finish it.
20 robustness probe (trailing silence) Ich gehe oft ins Kino. Read this sentence normally, then stay silent for a few seconds afterward before stopping, like you trailed off without saying anything more.
21 robustness probe (trailing silence) Er nimmt das Buch. Read this sentence normally, then stay silent for a few seconds afterward before stopping, like you trailed off without saying anything more.
22 disfluency (slow) Wir sitzen auf dem Balkon. Read this slowly and deliberately, like you’re carefully working out each word as a learner would.
23 disfluency Die äh Kinder haben viel Lärm gemacht. Read this naturally, including the filler word exactly as written, like a real hesitation.
24 disfluency (pause) Übermorgen fahre ich nach Hamburg. Read this sentence, but pause for a few seconds partway through, like you’re thinking mid-sentence, then finish it.
25 disfluency (pause) Wir haben einen langen Spaziergang gemacht. Read this sentence, but pause for a few seconds partway through, like you’re thinking mid-sentence, then finish it.
26 disfluency Die Tre— Treppe ist steil. Read this naturally, including the self-correction exactly as written — the em-dash (—) marks where you cut yourself off and restart the word right after it.
27 disfluency (pause) Das Buch ist sehr interessant. Read this sentence, but pause for a few seconds partway through, like you’re thinking mid-sentence, then finish it.
28 disfluency Wir sind gestern ins Ki— Kino gegangen. Read this naturally, including the self-correction exactly as written — the em-dash (—) marks where you cut yourself off and restart the word right after it.
29 disfluency Ich brauche ähm einen neuen Schlüssel. Read this naturally, including the filler word exactly as written, like a real hesitation.
30 disfluency (pause) Das Restaurant ist hier in der Nähe. Read this sentence, but pause for a few seconds partway through, like you’re thinking mid-sentence, then finish it.
31 disfluency Wie heißen ähm Si— Sie? — Ich heiße Müller. Read this naturally, including both the filler word and the self-correction exactly as written — the em-dash (—) marks where you cut yourself off and restart the word right after it.
32 disfluency (slow) Er hat eine gute Note. Read this slowly and deliberately, like you’re carefully working out each word as a learner would.
33 disfluency Ich äh mache meine Hausaufgaben. Read this naturally, including the filler word exactly as written, like a real hesitation.
34 disfluency Mittwoch ist äh mein fre— freier Tag. Read this naturally, including both the filler word and the self-correction exactly as written — the em-dash (—) marks where you cut yourself off and restart the word right after it.
35 disfluency (pause) Ich würde gern wissen, ob die Heizkosten schon dabei sind. Read this sentence, but pause for a few seconds partway through, like you’re thinking mid-sentence, then finish it.
36 disfluency Ich ähm lerne Deutsch, weil es mir Spaß macht. Read this naturally, including the filler word exactly as written, like a real hesitation.
37 disfluency Wir sind äh in der Schweiz. Read this naturally, including the filler word exactly as written, like a real hesitation.
38 disfluency Damals äh war das Leben anders — es gab zum Beis— Beispiel noch kein Internet. Read this naturally, including both the filler word and the self-correction exactly as written — the em-dash (—) marks where you cut yourself off and restart the word right after it.
39 disfluency Ich benutze den Staubsauger jed— jeden Samstag. Read this naturally, including the self-correction exactly as written — the em-dash (—) marks where you cut yourself off and restart the word right after it.
40 disfluency Das Buch ist se— sehr spannend. Read this naturally, including the self-correction exactly as written — the em-dash (—) marks where you cut yourself off and restart the word right after it.
41 disfluency Ich bewerbe äh mich um die Stelle im Marketing. Read this naturally, including the filler word exactly as written, like a real hesitation.
42 disfluency (pause) Können Sie die wichtigsten Punkte kurz zusammenfassen? Read this sentence, but pause for a few seconds partway through, like you’re thinking mid-sentence, then finish it.
43 disfluency (slow) Es war kalt, trotzdem gingen wir spazieren. Read this slowly and deliberately, like you’re carefully working out each word as a learner would.
44 disfluency Die Symptome äh sind Fieber, Husten und Halsschmerzen. Read this naturally, including the filler word exactly as written, like a real hesitation.
45 plain Trotz des schlechten Wetters gehen wir raus. Read this sentence naturally.
46 plain Die Epistemologie untersucht, was Wissen ist und wie wir es erlangen. Read this sentence naturally.
47 plain Die Verfassung garantiert die Meinungsfreiheit aller Bürger. Read this sentence naturally.
48 plain Wir müssen uns der Realität stellen, auch wenn sie unangenehm ist. Read this sentence naturally.