fixing some typos and bolding of example
Browse files
README.md
CHANGED
@@ -127,54 +127,55 @@ The model demonstrates high efficacy in correcting short, erroneous sentences, w
|
|
127 |
|
128 |
It performs nearly on par with GPTTurbo16k at around 93% sentence similarity. But there are gaps.
|
129 |
|
130 |
-
Take for example this output.
|
131 |
-
|
132 |
-
|
133 |
-
|
134 |
-
|
135 |
-
|
136 |
-
|
137 |
-
|
138 |
-
|
139 |
-
|
140 |
-
|
141 |
-
|
142 |
-
|
143 |
-
|
144 |
-
|
145 |
-
|
146 |
-
|
147 |
-
|
148 |
-
|
149 |
-
|
150 |
-
|
151 |
-
|
152 |
-
|
153 |
-
|
154 |
-
|
155 |
-
|
156 |
-
|
157 |
-
|
158 |
-
|
159 |
-
|
160 |
-
|
161 |
-
|
162 |
-
|
163 |
-
|
164 |
-
|
165 |
-
|
166 |
-
|
167 |
-
|
168 |
-
|
169 |
-
|
170 |
-
|
171 |
-
|
172 |
-
|
173 |
-
|
174 |
-
|
175 |
-
|
176 |
-
|
177 |
-
|
|
|
178 |
|
179 |
|
180 |
|
@@ -197,4 +198,5 @@ The model follows the T5 architecture, fine-tuned for the specific task of text
|
|
197 |
author={Will Wade},
|
198 |
year={2021},
|
199 |
howpublished={\url{https://huggingface.co/willwade/t5-small-spoken-typo}},
|
200 |
-
}
|
|
|
|
127 |
|
128 |
It performs nearly on par with GPTTurbo16k at around 93% sentence similarity. But there are gaps.
|
129 |
|
130 |
+
Take for example this output and bolded elements for parts that are incorrect.
|
131 |
+
|
132 |
+
|
133 |
+
Original: Didyoucatchthegamelastnight?
|
134 |
+
Corrected: Did you catch the game last night?
|
135 |
+
|
136 |
+
Original: Wannagrabcoffeetomorrow?
|
137 |
+
Corrected: Wanna grab coffee tomorrow?
|
138 |
+
|
139 |
+
Original: ImdyingsomeonecancellsoIcandogsitter!
|
140 |
+
Corrected: I'm dying someone **cancell** so I can dogsitter!
|
141 |
+
|
142 |
+
Original: Hahahahahahahathats hilarious!
|
143 |
+
Corrected: Haha ha ha ha that's hilarious!
|
144 |
+
|
145 |
+
Original: OMGyouneedtoseethelatestmeme!
|
146 |
+
Corrected: OMG you need to see the latest meme!
|
147 |
+
|
148 |
+
Original: Seriouslythisweatherissocrazy!
|
149 |
+
Corrected: Seriously this weather is so crazy!
|
150 |
+
|
151 |
+
Original: Whatchauptomefriend?
|
152 |
+
Corrected: What's **his** friend?
|
153 |
+
|
154 |
+
Original: Feelingburntoutaftettodayhelp!
|
155 |
+
Corrected: Feeling burnt out today help!
|
156 |
+
|
157 |
+
Original: Guesswhosingleagain!
|
158 |
+
Corrected: Guess who single again!
|
159 |
+
|
160 |
+
Original: Youwontyoubelievewhatjusthappened!
|
161 |
+
Corrected: You **want** you believe what just happened!
|
162 |
+
|
163 |
+
Original: Moviemarathonatmyplacethisweekend?
|
164 |
+
Corrected: Movie Marathon at my place this weekend?
|
165 |
+
|
166 |
+
Original: Needstudymotivationanyideas?
|
167 |
+
Corrected: Need study motivation any ideas?
|
168 |
+
|
169 |
+
Original: Sostressedaboutthispresentation!
|
170 |
+
Corrected: So stressed about this presentation!
|
171 |
+
|
172 |
+
Original: Finallyfinishedthatbookyourecommended!
|
173 |
+
Corrected: Finally finished that book **you're** recommended!
|
174 |
+
|
175 |
+
Original: Anygoodshowsbingeablelately?
|
176 |
+
Corrected: Any good shows **biteable** recently?
|
177 |
+
|
178 |
+
We hope to build on this by further fine-tuning in time
|
179 |
|
180 |
|
181 |
|
|
|
198 |
author={Will Wade},
|
199 |
year={2021},
|
200 |
howpublished={\url{https://huggingface.co/willwade/t5-small-spoken-typo}},
|
201 |
+
}
|
202 |
+
```
|