NEXUS IELTS SPEAKING PART AUDIO -> TIMESTAMP FRAME

Use this prompt in Google AI Studio with ONE IELTS Speaking examiner audio file at a time.

Run it separately for:

- Part 1 audio
- Part 3 audio

Do not use this prompt for Part 2. Part 2 is a visible cue card and has a separate normaliser prompt.

GOAL

Create a timestamp frame for NEXUS Test Importer -> IELTS Speaking -> Plain text -> tick Audio parser.

The output will be pasted into either:

- Part 1 audio parser text
- Part 3 audio parser text

BEFORE YOU START

Identify whether the uploaded audio is Part 1 or Part 3 from the file name, transcript, or my message.

If the part is unclear, write this only:

QA_NOTES
- BLOCKED: The uploaded audio part is unclear. Please tell me whether this is Part 1 or Part 3.

STRICT SOURCE RULES

- Use the uploaded audio as the timestamp source.
- If a PDF/image/source text is also uploaded, use it only to confirm wording.
- Do not invent questions.
- Do not invent missing audio.
- Do not estimate timestamps from reading speed.
- Do not include student answer time in any timestamp.
- If there is no audio file, do not produce Q lines.
- If the audio is unclear, mark the affected line under QA_NOTES.

TIMESTAMP RULES

- Timestamp only the examiner's actual prompt/question.
- Start time = when the examiner begins the prompt/question.
- End time = when the examiner finishes that prompt/question.
- Use timestamps relative to the uploaded part audio. Each part starts at 00:00.
- Use MM:SS or HH:MM:SS.
- Keep leading zeroes.
- Every Q line must have Q number, start-end timestamp, and exact question text.

OUTPUT FORMAT FOR PART 1 AUDIO

PART 1: AUDIO PARSER TEXT
Q1 00:00-00:08 [exact examiner prompt/question]
Q2 00:15-00:18 [exact examiner prompt/question]
Q3 00:25-00:35 [exact examiner prompt/question]

QA_NOTES
- [Only real uncertainty, background noise, unclear wording, mismatch with source text, or missing audio region.]

OUTPUT FORMAT FOR PART 3 AUDIO

PART 3: AUDIO PARSER TEXT
Q1 00:00-00:08 [exact examiner question]
Q2 00:12-00:20 [exact examiner question]
Q3 00:25-00:34 [exact examiner question]

QA_NOTES
- [Only real uncertainty, background noise, unclear wording, mismatch with source text, or missing audio region.]

FINAL CHECK BEFORE YOU ANSWER

- Output only one part frame: either Part 1 or Part 3.
- Do not output TITLE, TARGET BAND, TIMING, or Part 2.
- Do not output placeholder cue cards.
- Do not write "estimated" timestamps. If you cannot hear it, block it in QA_NOTES.
- Do not include commentary outside the required output blocks.