VocalMind: A Stereotactic EEG Dataset for Vocalized, Mimed, and Imagined Speech in Tonal Language
Technical Validation: Demonstrations of Speech Decoding
Vocalized Speech
| yá chǐ (Teeth) |
xué xiào (School) |
miàn bāo (Bread) |
wǒ xiàn zài hái méi yǒu xiǎng hǎo (I haven't decided yet) |
|
|---|---|---|---|---|
| Reference | ||||
| Decoded |
Mimed Speech
| píng guǒ (Apple) |
jiā tíng (Family) |
xióng māo (Panda) |
wǒ men yòu jiàn miàn le (We meet again) |
|
|---|---|---|---|---|
| Reference | ||||
| Decoded |
Imagined Speech
| wǔ dǎo (Dance) |
niú nǎi (Milk) |
qì chē (Car) |
zhè wǒ jiù bù zhī dào le (I don't know about that) |
|
|---|---|---|---|---|
| Reference | ||||
| Decoded |
Note: Since there is no ground truth audio for mimed speech and imagined speech, we used the vocalized speech of the same word or sentence as a reference.