mirrors/gst-plugins-rs
mirrors/gst-plugins-rs
1
1
Fork 0
mirror of https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs.git synced 2025-01-10 19:25:26 +00:00
Code Issues Projects Releases Wiki Activity

awstranscriber: fix what we send over for translations

Browse source 
Prior to this commit, we were sending over words concatenated together
with no separators, for instance "Idon'twanttobeanemperor".

The translation service seems clever enough to translate the contents
anyway, but there is no reason to make its task harder than necessary,
and it didn't re-add separators when the target language was the same as
the source language, which resulted in less than ideal output.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1171>
This commit is contained in:
Mathieu Duponchelle 2023-04-10 16:34:49 +02:00 committed by GStreamer Marge Bot
parent 408fd2030c
commit f366c20869
1 changed files with 28 additions and 17 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

45
net/aws/src/transcriber/translate.rs
View file

@ -109,23 +109,34 @@ impl TranslateLoop {
continue;
}
let (ts_duration_list, content): (Vec<(gst::ClockTime, gst::ClockTime)>, String) =
transcript_items
.iter()
.map(|item| {
(
(item.pts, item.duration),
match self.tokenization_method {
Tokenization::None => item.content.clone(),
Tokenization::SpanBased => {
format!("{SPAN_START}{}{SPAN_END}", item.content)
}
},
)
})
.unzip();
let mut ts_duration_list: Vec<(gst::ClockTime, gst::ClockTime)> = vec![];
let mut content: Vec<String> = vec![];
gst::trace!(CAT, imp: self.pad, "Translating {content} with {ts_duration_list:?}");
let mut it = transcript_items.iter().peekable();
while let Some(item) = it.next() {
let suffix = match it.peek() {
Some(next_item) => {
if next_item.is_punctuation {
""
} else {
" "
}
}
None => "",
};
ts_duration_list.push((item.pts, item.duration));
content.push(match self.tokenization_method {
Tokenization::None => format!("{}{}", item.content, suffix),
Tokenization::SpanBased => {
format!("{SPAN_START}{}{SPAN_END}{}", item.content, suffix)
}
});
}
let content: String = content.join("");
gst::debug!(CAT, imp: self.pad, "Translating {content} with {ts_duration_list:?}");
let translated_text = self
.client
@ -143,7 +154,7 @@ impl TranslateLoop {
.translated_text
.unwrap_or_default();
gst::trace!(CAT, imp: self.pad, "Got translation {translated_text}");
gst::debug!(CAT, imp: self.pad, "Got translation {translated_text}");
let translated_items = match self.tokenization_method {
Tokenization::None => {