All labs / Audio Checking model…
03.1 — Audio / Beginner

Voice

Audio Beginner Mixed

Two models pointed in opposite directions. Synthesise a voice with adjustable rate and pitch, then feed real audio to a transcriber and see where it mishears.

Parameters
39M · Whisper tiny; TTS voices are a hosted service
Input
Text or audio
Output
Audio or text
Typical time
1–5s
Last run
Text to speak 0 / 1000
Output

Press speak. The waveform above is decoded from the actual audio — the tall sections are vowels, the narrow spikes are consonants, and the flat gaps are the pauses the model chose to insert.

How it works

Transcription converts sound into a spectrogram — a picture of which frequencies are loud over time — then treats reading that picture as a translation problem, emitting text tokens. Synthesis runs the pipeline in reverse.

Full identifier
openai/whisper-tiny (STT) · Microsoft Edge neural voices (TTS)
Where it runs
Partly local, partly hosted
Parameters
39M

Recognition accuracy varies substantially across accents and dialects, reflecting what was in the training audio.

What to notice
  • Whisper tiny is the smallest of six sizes. It trades accuracy for speed, and it shows on accents, names and technical vocabulary.
  • It never returns an empty transcript. Given silence or noise it will produce plausible words anyway.
  • Synthesised prosody comes from punctuation. Move a comma and the phrasing of the sentence changes.
A note on language

This page says “the model predicts”, not “the model knows”. That is deliberate. None of these systems understand the images or sentences you give them; they map inputs to outputs using patterns fixed at training time. The difference matters most exactly when the output is impressive.