Archives
Archives
| From: | wabe <wabenbau@×××××.com> | ||
|---|---|---|---|
| To: | gentoo-user@l.g.o | ||
| Subject: | Re: [gentoo-user] speech recognition? | ||
| Date: | Wed, 18 May 2016 01:41:19 | ||
| Message-Id: | 20160518034043.67d74828@hal9000.localdomain | ||
| In Reply to: | Re: [gentoo-user] speech recognition? by Andrew Savchenko |
||
| 1 | Andrew Savchenko <bircoph@g.o> wrote: |
| 2 | |
| 3 | > > https://cloud.google.com/speech/ |
| 4 | > |
| 5 | > Sad, but there are no free software solutions available in this |
| 6 | > area; at least I'm not aware of them completely. This is |
| 7 | > understandable: the task is extraordinary in both manpower and |
| 8 | > computational resources required. |
| 9 | |
| 10 | Actually there are some open source speech recognition programs. |
| 11 | |
| 12 | https://en.wikipedia.org/wiki/List_of_speech_recognition_software |
| 13 | |
| 14 | Sphinx2 and Sphinx3 are even available as gentoo ebuilds |
| 15 | (app-accessibility/sphinx2 and app-accessibility/sphinx3). |
| 16 | |
| 17 | Sphinx4 as well as a lots of information about it is available |
| 18 | here: |
| 19 | |
| 20 | http://cmusphinx.sourceforge.net/ |
| 21 | |
| 22 | A German language model can be found here: |
| 23 | |
| 24 | https://sourceforge.net/projects/cmusphinx/files/Acoustic%20and%20Language%20Models/German%20Voxforge/ |
| 25 | |
| 26 | However I don't know how much effort will be necessary to get the |
| 27 | whole thing working for your purposes. |
| 28 | |
| 29 | |
| 30 | > And what Google offers here is not even a proprietary application, |
| 31 | > but an access to a proprietary cloud service which is temporarily |
| 32 | > free of charge as long as it is in alpha-testing stage. |
| 33 | |
| 34 | That's right. Beside the fact that every spoken word will be |
| 35 | transfered to Google (bad enough), it will not be free of charge for |
| 36 | ever. |
| 37 | |
| 38 | -- |
| 39 | Regards |
| 40 | wabe |