RapidEars ignoring secondsOfSilenceToDetect – Politepix /forums/topic/rapidears-ignoring-secondsofsilencetodetect/feed/ Tue, 23 Apr 2024 14:42:50 +0000 https://bbpress.org/?v=2.6.9 en-US /forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020739 <![CDATA[RapidEars ignoring secondsOfSilenceToDetect]]> /forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020739 Tue, 08 Apr 2014 01:16:08 +0000 morchella It appears that when setFinalizeHypothesis = FALSE, RapidEars ignores the setting for secondsOfSilenceToDetect. Is this a bug? I don’t need the finalized hypothesis, but I do need to control how long pocketsphinxController waits before considering the utterance ended.

]]>
/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020742 <![CDATA[Reply To: RapidEars ignoring secondsOfSilenceToDetect]]> /forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020742 Tue, 08 Apr 2014 02:27:54 +0000 Halle Winkler Hello,

RapidEars doesn’t use secondsOfSilenceToDetect. Both live and finalized hypotheses are delivered as soon as they are available.

]]>
/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020743 <![CDATA[Reply To: RapidEars ignoring secondsOfSilenceToDetect]]> /forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020743 Tue, 08 Apr 2014 02:44:14 +0000 morchella Sorry if I was unclear. There is a rapidEarsDidDetectEndOfSpeech delegate method which appears to always be called 50-300ms after pocketsphinxDidDetectFinishedSpeech. I have observed that these end of speech callbacks are sensitive to secondsOfSilenceToDetect, but only when setFinalizeHypothesis = TRUE.

Is this the intended behavior?

]]>
/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020744 <![CDATA[Reply To: RapidEars ignoring secondsOfSilenceToDetect]]> /forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020744 Tue, 08 Apr 2014 02:59:30 +0000 Halle Winkler That’s correct – finalizing the hypothesis means that the end of speech is detected and reported, while using live recognition only means that there is no wait for a pause or consequent callback when a pause is detected because a pause at the end of an utterance isn’t used by the engine for determining the hypothesis.

]]>
/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020745 <![CDATA[Reply To: RapidEars ignoring secondsOfSilenceToDetect]]> /forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020745 Tue, 08 Apr 2014 03:25:40 +0000 morchella Yes, but even when setFinalizeHypothesis = FALSE, the end of speech is still detected and reported. It’s just that the config option (secondsOfSilenceToDetect) is now ignored.

I can work around it, but it seems like incorrect behavior.

]]>
/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020749 <![CDATA[Reply To: RapidEars ignoring secondsOfSilenceToDetect]]> /forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020749 Tue, 08 Apr 2014 07:25:14 +0000 Halle Winkler I hear what you’re saying, but it is by design. Live mode does not do any kind of waiting for a pause in order to derive state, so if the engine is only operating in live mode, it uses its own logic to determine when is a good time to call an utterance over so that continuous recognition is able to proceed without notable pauses or skips. secondsOfSilenceToDetect and rapidEarsDidDetectEndOfSpeech both refer to pause detection, which isn’t a feature of live mode. It could be documented better, I agree.

I think that if I wanted to use live mode hypotheses and non-live mode utterance logic I’d probably turn finalize on and just ignore its hyp output. The overhead isn’t that heavy.

]]>
/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020764 <![CDATA[Reply To: RapidEars ignoring secondsOfSilenceToDetect]]> /forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020764 Tue, 08 Apr 2014 16:41:43 +0000 morchella Okay, thanks Halle, I appreciate the advice!

]]>
This XML file does not appear to have any style information associated with it. The document tree is shown below.
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:wfw="http://wellformedweb.org/CommentAPI/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0">
<channel>
<title>RapidEars ignoring secondsOfSilenceToDetect – Politepix</title>
<atom:link href="/forums/topic/rapidears-ignoring-secondsofsilencetodetect/feed/" rel="self" type="application/rss+xml"/>
<link>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/feed/</link>
<description/>
<lastBuildDate>Tue, 23 Apr 2024 14:42:50 +0000</lastBuildDate>
<generator>https://bbpress.org/?v=2.6.9</generator>
<language>en-US</language>
<item>
<guid>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020739</guid>
<title>
<![CDATA[ RapidEars ignoring secondsOfSilenceToDetect ]]>
</title>
<link>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020739</link>
<pubDate>Tue, 08 Apr 2014 01:16:08 +0000</pubDate>
<dc:creator>morchella</dc:creator>
<description>
<![CDATA[ <p>It appears that when setFinalizeHypothesis = FALSE, RapidEars ignores the setting for secondsOfSilenceToDetect. Is this a bug? I don&#8217;t need the finalized hypothesis, but I do need to control how long pocketsphinxController waits before considering the utterance ended.</p> ]]>
</description>
</item>
<item>
<guid>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020742</guid>
<title>
<![CDATA[ Reply To: RapidEars ignoring secondsOfSilenceToDetect ]]>
</title>
<link>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020742</link>
<pubDate>Tue, 08 Apr 2014 02:27:54 +0000</pubDate>
<dc:creator>Halle Winkler</dc:creator>
<description>
<![CDATA[ <p>Hello,</p> <p>RapidEars doesn&#8217;t use secondsOfSilenceToDetect. Both live and finalized hypotheses are delivered as soon as they are available.</p> ]]>
</description>
</item>
<item>
<guid>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020743</guid>
<title>
<![CDATA[ Reply To: RapidEars ignoring secondsOfSilenceToDetect ]]>
</title>
<link>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020743</link>
<pubDate>Tue, 08 Apr 2014 02:44:14 +0000</pubDate>
<dc:creator>morchella</dc:creator>
<description>
<![CDATA[ <p>Sorry if I was unclear. There is a rapidEarsDidDetectEndOfSpeech delegate method which appears to always be called 50-300ms after pocketsphinxDidDetectFinishedSpeech. I have observed that these end of speech callbacks are sensitive to secondsOfSilenceToDetect, but only when setFinalizeHypothesis = TRUE.</p> <p>Is this the intended behavior?</p> ]]>
</description>
</item>
<item>
<guid>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020744</guid>
<title>
<![CDATA[ Reply To: RapidEars ignoring secondsOfSilenceToDetect ]]>
</title>
<link>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020744</link>
<pubDate>Tue, 08 Apr 2014 02:59:30 +0000</pubDate>
<dc:creator>Halle Winkler</dc:creator>
<description>
<![CDATA[ <p>That&#8217;s correct – finalizing the hypothesis means that the end of speech is detected and reported, while using live recognition only means that there is no wait for a pause or consequent callback when a pause is detected because a pause at the end of an utterance isn&#8217;t used by the engine for determining the hypothesis.</p> ]]>
</description>
</item>
<item>
<guid>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020745</guid>
<title>
<![CDATA[ Reply To: RapidEars ignoring secondsOfSilenceToDetect ]]>
</title>
<link>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020745</link>
<pubDate>Tue, 08 Apr 2014 03:25:40 +0000</pubDate>
<dc:creator>morchella</dc:creator>
<description>
<![CDATA[ <p>Yes, but even when setFinalizeHypothesis = FALSE, the end of speech is still detected and reported. It&#8217;s just that the config option (secondsOfSilenceToDetect) is now ignored.</p> <p>I can work around it, but it seems like incorrect behavior.</p> ]]>
</description>
</item>
<item>
<guid>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020749</guid>
<title>
<![CDATA[ Reply To: RapidEars ignoring secondsOfSilenceToDetect ]]>
</title>
<link>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020749</link>
<pubDate>Tue, 08 Apr 2014 07:25:14 +0000</pubDate>
<dc:creator>Halle Winkler</dc:creator>
<description>
<![CDATA[ <p>I hear what you&#8217;re saying, but it is by design. Live mode does not do any kind of waiting for a pause in order to derive state, so if the engine is only operating in live mode, it uses its own logic to determine when is a good time to call an utterance over so that continuous recognition is able to proceed without notable pauses or skips. secondsOfSilenceToDetect and rapidEarsDidDetectEndOfSpeech both refer to pause detection, which isn&#8217;t a feature of live mode. It could be documented better, I agree.</p> <p>I think that if I wanted to use live mode hypotheses and non-live mode utterance logic I&#8217;d probably turn finalize on and just ignore its hyp output. The overhead isn&#8217;t that heavy.</p> ]]>
</description>
</item>
<item>
<guid>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020764</guid>
<title>
<![CDATA[ Reply To: RapidEars ignoring secondsOfSilenceToDetect ]]>
</title>
<link>/forums/topic/rapidears-ignoring-secondsofsilencetodetect/#post-1020764</link>
<pubDate>Tue, 08 Apr 2014 16:41:43 +0000</pubDate>
<dc:creator>morchella</dc:creator>
<description>
<![CDATA[ <p>Okay, thanks Halle, I appreciate the advice!</p> ]]>
</description>
</item>
</channel>
</rss>