摘要 |
<p>In a speech recognition method for activating a hyperlink (HL) of text information (TI) of an Internet page (8) displayed for a user, the method comprising the following steps: determining hypertexts (HT(HL)) of the hyperlinks (HL) in the text information (TI) shown and determining corresponding first phoneme sequences (PI1[HT(HL)]) of these hypertexts (HT(HL)) and receiving a spoken command (AS) from a user which represents at least part of a determined hypertext (HT(HL)), and determining a second phoneme sequence (PI2) corresponding to the spoken command (AS) and determining the hyperlink (HL) selected by the user by comparison of the determined second phoneme sequence (PI2) with determined first phoneme sequences (PI1[HT(HL)]) and activating the selected hyperlink (HL), where one quality value (QV) is determined for each hypertext (HT(HL)) when first phoneme sequences (PI1[(HT(HL)]) corresponding to the hypertexts (HT(HL)) are determined and where an extra hypertext (ZHT(HL)) is determined when the quality value (QV) of a hypertext (HT(HLL)) falls short of a quality threshold (QS) and is assigned to the hypertext (HT(HL)) in the text information (TI) of the Internet page (25, 26) or in lieu of the hypertext (HT(HL)), and where a first phoneme sequence (PI1[HT(HL)]) determined for the extra hypertext (ZHT(HL)) has a quality value (QV) that exceeds the quality threshold (QS).</p> |