PTE Speaking Practice (Timed Drills with Feedback)

About the PTE Speaking Practice (Timed Drills with Feedback)

Practise the speaking part of PTE Academic the way the test runs it: the same eight task types, Pearson’s preparation and answer times, audio that plays once, and a microphone that closes after three seconds of silence. Every practice item — texts, sentences, charts, lectures, discussions and situations — was written for MySmartCoPilot; none is copied from Pearson or from past tests.

After each answer you can play your recording back and see when you started, how long you spoke and where you paused. Turn on transcription and your answer is transcribed on your own device, never on a server — by your browser’s own speech recognition when English is installed for it on the device, otherwise by the Whisper speech model — to show what it heard: in Read Aloud and Repeat Sentence every word you replaced, missed or added; in Describe Image, Retell Lecture, Summarize Group Discussion and Respond to a Situation the key points you covered; in Answer Short Question whether your answer was right; and your pace, hesitations and repeats.

This is practice feedback, not a score: MySmartCoPilot never predicts a PTE score, and it is not affiliated with Pearson.

The speaking tasks and their timings

• Personal Introduction: 25 seconds to prepare, 30 seconds to speak. Not scored: Pearson says it is for familiarisation only.
• Read Aloud (6–7 in a test): a text of up to 60 words; 30–40 seconds to prepare, then a short tone; the reading time varies with the length of the text. MySmartCoPilot gives 30, 35 or 40 seconds to prepare and 35 or 40 seconds to read, by the length of the text.
• Repeat Sentence (10–12): a sentence of 3–9 seconds, played once; 15 seconds to repeat it after a short tone.
• Describe Image (5–6): 25 seconds to study the image, 40 seconds to describe it.
• Retell Lecture (2–3): the image appears 3 seconds before a lecture of up to 90 seconds; then 10 seconds to prepare and 40 seconds to retell it.
• Answer Short Question (5–6): a question of 3–9 seconds, played once; 10 seconds to answer in a word or a few words, with no tone.
• Summarize Group Discussion (2–3): a discussion between 3 people of up to 3 minutes; 10 seconds to prepare and 2 minutes to summarise it.
• Respond to a Situation (2–3): a situation of up to 60 words, shown and played once; 10 seconds to prepare and 40 seconds to respond.

Timings and rules come from Pearson’s PTE Academic test format page; the number of items of each type and the traits that are scored come from the PTE Academic Test Taker Score Guide. Check both before your test: Pearson can change the format.

The three-second rule

Pearson’s test format page says that in Answer Short Question, and also in Read Aloud, Repeat Sentence, Describe Image and Retell Lecture, the recording ends if you wait more than three seconds before speaking, and if you stop for three seconds while speaking you cannot continue. The tool follows the level of your microphone and closes it in the same way (silence here means the level has dropped back to near the room’s background noise, measured while you prepare). The pause timeline marks where the test would have closed the microphone, so you can see whether a hesitation would have cost you the rest of the answer.

What the feedback measures

• From the recording itself (always): when you started, your speaking time, and every silence of a quarter of a second or more inside the answer; silences of a second or more count as long pauses.
• From the transcript (when transcription is on): words a minute from your first to your last word, hesitation sounds and filler phrases the transcript shows, repeated words, and the content check of each task.
• Content checks: Read Aloud and Repeat Sentence compare the words you said with the text, word by word, after folding spelling and number differences (“colour” and “color”, “25” and “twenty-five”); a word heard as a similar-sounding one (“sheep” for “ship”) is flagged as a pronunciation hint. Describe Image, Retell Lecture, Summarize Group Discussion and Respond to a Situation are checked against a list of key points written for each item, and lectures and discussions also show how much of your answer repeats the source word for word — Pearson’s descriptors reward ideas in your own words.
• What is not measured: pronunciation quality and oral fluency on Pearson’s 0–5 scales, and anything else that needs an examiner or Pearson’s own scoring engine. MySmartCoPilot gives no PTE score, predicted or otherwise.

Limitations

• Practice feedback only: there is no official or predicted PTE score, and the content checks are keyword and word-matching checks, not Pearson’s scoring.
• Speech recognition can be wrong — with names, noise, quiet speech or a strong accent — and it often leaves out “um” and “uh”; read the transcript before you trust a missed word.
• Audio prompts use your browser’s synthetic voices, which sound different from the recorded speakers in the test. Browsers without an English voice show the text instead.
• Transcription needs your browser’s on-device English recognition or a one-time model download, and a reasonably recent device; with the model, a long answer can take tens of seconds to transcribe on a slow phone.
• MySmartCoPilot is not affiliated with or endorsed by Pearson. All practice items are MySmartCoPilot’s own; real test items will differ.

Privacy

The microphone is used only while an item runs. Recordings stay in this page’s memory: they are not saved, not uploaded and gone when you leave the page (download one with the Recording button if you want to keep it). Transcripts are made on your device: by your browser’s own speech recognition when it reports that English is installed on this device (never its online service), otherwise by Whisper through MySmartCoPilot’s on-device AI, downloaded once after you agree. There is no server option in this tool. Your progress table (task, item, content share, pace and pauses — no audio and no transcript) is kept in this browser until you clear it. Audio prompts are read by your browser’s own voices; if your browser only has online voices, it sends MySmartCoPilot’s practice text — never your recording — to its maker to be spoken.