Received: from mail-qk0-f191.google.com ([209.85.220.191]:33128) by stodi.digitalkingdom.org with esmtps (TLSv1.2:AES128-GCM-SHA256:128) (Exim 4.85) (envelope-from ) id 1Z8su3-0001nS-7V for lojban-list-archive@lojban.org; Sat, 27 Jun 2015 09:24:32 -0700 Received: by qkbx78 with SMTP id x78sf33319223qkb.0 for ; Sat, 27 Jun 2015 09:24:21 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlegroups.com; s=20120806; h=date:from:to:cc:message-id:in-reply-to:references:subject :mime-version:content-type:x-original-sender:reply-to:precedence :mailing-list:list-id:x-spam-checked-in-group:list-post:list-help :list-archive:sender:list-subscribe:list-unsubscribe; bh=oveoUo4YagouLc3ONpy2eGdFGUtbEB3qsk406vsXvoo=; b=nV0qNlrxY1Q3PkTDFxr4Wua6bkuokCIF5mSnXjJAB0jkJ74VaVokwLx+ldF28pnGoU HdjTQkXgNLl9ao8IWETfxQKj/Mb57+IM8a8CBiUz2HicsknOpTo2WlcAsc+pkohVfjVM W5+aebgpWvvwFIFuapZC4Kp31ZPzKwseRjbAHOo6uw+aBPL5PkoQjAE/iNl4hrwej8Xh yKYM2sncTKpLkcQDJnEhhUrpMmcZCBFC/G9mwo/HXAbuo3TBU9LCyRO9TJWdaHp+lnPc llOM1gxCXZDh5ITMzwDVIu31ArmtpW/wHCwbLOM85rYzL6Cxa67dbBQc4kC7V/KSaLY/ pXwQ== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=date:from:to:cc:message-id:in-reply-to:references:subject :mime-version:content-type:x-original-sender:reply-to:precedence :mailing-list:list-id:x-spam-checked-in-group:list-post:list-help :list-archive:sender:list-subscribe:list-unsubscribe; bh=oveoUo4YagouLc3ONpy2eGdFGUtbEB3qsk406vsXvoo=; b=tZHhyBXX42V9pAIbEHCe4zrk11gEtI45znIxmFhr3OHeynI5n3sR5O/k21PqczQ8XG J8b+RrhLFKJUmv9nEaWu5O2lxj+ne7G0/SITCchTSXEqWdY2R1rDMh5S/IrruDqm4wE5 4in3AsxRCwABafU0zrPhvPFDyjmOrMWXcxnvY2KP2DZHjjCqlRR9RwH0vjzmVDSHDHxC us4HLy9kfoxS5Hx5AE3l86mVRFF/+9L2Jj2NpVz3IoqrW2agqd+TnBdM5FNdHgZjOezq mfJi4IAB2Hdtyi7FSkTyPsP37SvHxZx+78dvoQTXjpRrUFs20QLIpowIaZHiZ12atoDa 8ljQ== X-Received: by 10.140.82.166 with SMTP id h35mr93286qgd.36.1435422261154; Sat, 27 Jun 2015 09:24:21 -0700 (PDT) X-BeenThere: lojban@googlegroups.com Received: by 10.140.35.199 with SMTP id n65ls2593798qgn.21.gmail; Sat, 27 Jun 2015 09:24:20 -0700 (PDT) X-Received: by 10.140.38.180 with SMTP id t49mr92486qgt.9.1435422260832; Sat, 27 Jun 2015 09:24:20 -0700 (PDT) Date: Sat, 27 Jun 2015 09:24:20 -0700 (PDT) From: sorpa'as plat To: lojban@googlegroups.com Cc: timothy.lawrence@connect.qut.edu.au Message-Id: In-Reply-To: References: Subject: Re: [lojban] The Prototype of a Lojban Speech Recognition Tool MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_2937_1911882265.1435422260295" X-Original-Sender: sorpaas@gmail.com Reply-To: lojban@googlegroups.com Precedence: list Mailing-list: list lojban@googlegroups.com; contact lojban+owners@googlegroups.com List-ID: X-Spam-Checked-In-Group: lojban@googlegroups.com X-Google-Group-Id: 1004133512417 List-Post: , List-Help: , List-Archive: , List-Unsubscribe: , X-Spam-Score: -3.1 (---) X-Spam_score: -3.1 X-Spam_score_int: -30 X-Spam_bar: --- ------=_Part_2937_1911882265.1435422260295 Content-Type: multipart/alternative; boundary="----=_Part_2938_207417794.1435422260295" ------=_Part_2938_207417794.1435422260295 Content-Type: text/plain; charset=UTF-8 Hi Timothy, Thanks for the advice, but I think we may just stick on Sphinx for now. Sphinx uses hidden Markov acoustic models (HMMs) while Kaldi uses deep neural networks (DNNs). It's harder to train DNNs than to make use of HMMs, and we simply lack that manpower to do that :) Wei mu'o mi'e la sorpa'as On Saturday, June 27, 2015 at 2:21:03 AM UTC-4, Timothy Lawrence wrote: > > I really like this initiative :) > > > A friend of mine is working on speech recognition (for other purposes) > and tried Sphinx but ended up changing to > http://kaldi.sourceforge.net/about.html, I believe because he found it to > be better at recognising speech and/or it was faster. I thought I'd mention > it in case you're not fully set on Sphinx. > > ------------------------------ > *From:* loj...@googlegroups.com > on behalf of sorpa'as plat > > *Sent:* Saturday, 27 June 2015 1:29 PM > *To:* loj...@googlegroups.com > *Subject:* [lojban] The Prototype of a Lojban Speech Recognition Tool > > Hi all, > > I'm trying to build a Lojban speech recognition called tersku. Instead of > building an acoustic model by hand (which may need many manpower and takes > a long time), the attempt is to take the English acoustic model (which is > pretty mature) and adapt it for Lojban sounds. > > A running prototype can be found at https://git.null.tl/tersku.git (use *git://git.null.tl/tersku.git > * to clone). The prototype uses a > unmodified version of CMU's generic English acoustic model, with only > necessary dictionary and grammars to parse the text "le tanxe be le birka > cu cpana le tanxe be le botpi". To use it, recording a version of the text > "le tanxe be le birka cu cpana le tanxe be le botpi", convert the recording > to wav format, and replace the /resources/org/lojban/tersku/recording.wav > file with it. The program will output the best "hypothesis" for the text. > > The program does not work really well. That means there's lots of work and > I would appreciate your help. Below are some details of things to be done. > > *About the Program* > tersku uses CMU's Sphinx speech recognition engine. You can find Sphinx's > tutorials and documentations at http://cmusphinx.sourceforge.net. > > *Adapt the Acoustic Model* > The adaptation requires some 16KHZ single-channel wav recordings. Help are > appreciated if someone can create a Lojban phrase recording collection. > Note that a phrase recording collection will benefit the whole Lojban > community but not just the speech recognition program :) > > *Finish the Dictionary* > The dictionary in the prototype locates at > resources/org/lojban/tersku/jbo-1.dict. Because we are trying to adapt the > English acoustic model, all the phones are represented in Arpabet (*https://en.wikipedia.org/wiki/Arpabet > *). We will need to a) confirm > which arpabet symbol represents which Lojban sound, and b) write a program > that generates all the words in "[lojban word] [arpabet symbols]". This is > probably dependent of the adaptation of the acoustic model. > > *Finish the Grammar* > The grammar needs to be written in JSGF format ( > http://cmusphinx.sourceforge.net/wiki/tutoriallm). This haven't been > started yet (which needs help!). > > *Correct Me!* > There must be mistakes and errors both in the codes and in the recognition > details (I'm new to speech recognition!). > > Feel free to reach me at this email address or by opening an task at > https://phabricator.null.tl. I'm really looking forward to a Lojban > speech recognition tool, because it should be one of the features of Lojban > :) > > Wei > mu'o mi'e la sorpa'as > > -- > You received this message because you are subscribed to the Google Groups > "lojban" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to lojban+un...@googlegroups.com . > To post to this group, send email to loj...@googlegroups.com > . > Visit this group at http://groups.google.com/group/lojban. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "lojban" group. To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com. To post to this group, send email to lojban@googlegroups.com. Visit this group at http://groups.google.com/group/lojban. For more options, visit https://groups.google.com/d/optout. ------=_Part_2938_207417794.1435422260295 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Hi Timothy,

Thanks for the advice, but I think we m= ay just stick on Sphinx for now. Sphinx uses hidden Markov acoustic models = (HMMs) while Kaldi uses deep neural networks (DNNs). It's harder to train D= NNs than to make use of HMMs, and we simply lack that manpower to do that := )

Wei
mu'o mi'e la sorpa'as

On Saturday, June 27, 2015 at = 2:21:03 AM UTC-4, Timothy Lawrence wrote:

I really like this initiative :)


A friend of mine is working on speech recognition (for other purposes) a= nd tried Sphinx but ended up changing to http://kaldi.sourceforge.net/about.html, I believe because he foun= d it to be better at recognising speech and/or it was faster. I thought I'd= mention it in case you're not fully set on Sphinx.


From: loj...@googlegroups.com <loj...@googlegroups.com> on behalf of sorpa'= as plat <sor...@gm= ail.com>
Sent: Saturday, 27 June 2015 1:29 PM
To: loj...@go= oglegroups.com
Subject: [lojban] The Prototype of a Lojban Speech Recognition Tool<= /font>
 
Hi all,

I'm trying to build a Lojban speech recognition called tersku. Instead of b= uilding an acoustic model by hand (which may need many manpower and takes a= long time), the attempt is to take the English acoustic model (which is pr= etty mature) and adapt it for Lojban sounds.

A running prototype can be found at git.null.tl/tersku= .git to clone). The prototype uses a unmodified version of CMU's ge= neric English acoustic model, with only necessary dictionary and grammars t= o parse the text "le tanxe be le birka cu cpana le tanxe be le botpi". To u= se it, recording a version of the text "le tanxe be le birka cu cpana le tanxe be le botpi"= , convert the recording to wav format, and replace the /resources/org/lojba= n/tersku/recording.wav file with it. The program will output the best = "hypothesis" for the text.

The program does not work really well. That means there's lots of work and = I would appreciate your help. Below are some details of things to be done.<= br>
About the Program
tersku uses CMU's Sphinx speech recognition engine. You can find Sphinx's t= utorials and documentations at http://cmusphinx.sourc= eforge.net.

Adapt the Acoustic Model
The adaptation requires some 16KHZ single-channel wav recordings. Help are = appreciated if someone can create a Lojban phrase recording collection. Not= e that a phrase recording collection will benefit the whole Lojban communit= y but not just the speech recognition program :)

Finish the Dictionary
The dictionary in the prototype locates at resources/org/lojban/tersku/jbo-1.dict. Because we are trying to adapt the English acoustic model, all= the phones are represented in Arpabet (https://en.wikipedia.org/wiki/Arpabet). We will need to a) confirm which arpabet symbol represents which Lojban sound, and b) write a program= that generates all the words in "[lojban word] [arpabet symbols]". This is= probably dependent of the adaptation of the acoustic model.

Finish the Grammar
The grammar needs to be written in JSGF format (http://cmusphinx.sourceforge.net/wiki/tutoriallm). This hav= en't been started yet (which needs help!).

Correct Me!
There must be mistakes and errors both in the codes and in the recognition = details (I'm new to speech recognition!).

Feel free to reach me at this email address or by opening an task at https://phabricator.null.tl. I'm really looking forward to a Lojban spe= ech recognition tool, because it should be one of the features of Lojban :)=

Wei
mu'o mi'e la sorpa'as

--
You received this message because you are subscribed to the Google Groups "= lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an e= mail to lojban+un...@goo= glegroups.com.
To post to this group, send email to loj...@googlegroups.com.
Visit this group at http://groups.google.com/group/l= ojban.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups &= quot;lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an e= mail to lojban+unsub= scribe@googlegroups.com.
To post to this group, send email to lojban@googlegroups.com.
Visit this group at http:= //groups.google.com/group/lojban.
For more options, visit http= s://groups.google.com/d/optout.
------=_Part_2938_207417794.1435422260295-- ------=_Part_2937_1911882265.1435422260295--