OOV – Out of vocabulary – questfox is sometimes leaving us speechless.
Week by week we see incredible advances in the development of our speech-to-text approach inside of questfox. One of the everlasting challenges is the wrong transcription of words which are not correctly understood in a specific context.
If something is misspelled by a computer this normally happens when the word used is outside the standard vocabulary of the application. This tends to happen often when words are used in a company specific context. By the way this seems to be true for every single market research project.
When asking about a brand some may be asking for competitors with a very special name and a very special spelling. In these cases the transcription technology tends to fail to recognize those specific names. Technically speaking the words used are “out of vocabulary” (OOV).
The questfox solution to the OOV problem: It is now possible to create a list of special words to be integrated in questfox. Everytime the transcription engine of questfox is used this list is sent to the transcription engine to have a better understanding of specific vocabulary used.
Here is our questfox experiment to test the tool with some less common brands
|Brand||Transcription without OOV List||Transcription with OOV List|
|pangea labs||Pangea Labs||pangea LabsPangea Labs|
|Wisdom of Krauts||wisdom of crowds||wisdom of crowds|
|Tjaereborg Reisen||Sherlock Reisen|
|Comme des Garçons||komm DKVkomm digga Song||comme de garcon Gasthaus.com Digger|
kommen die ganzen
As you can see the overall transcription results do improve. But some issues remain.
Those who use mixed spoken language inside one dictation process may be familiar with the fact that transcription only works for one language at a time and hardly ever for two languages simultaneously.
As questfox works with an “expected language” tag for the transcription everything which is far away from the expected language is difficult to understand by the tool. In our pretest language German it was not possible to capture the french brand “Comme des Garçons”. We also failed with our made up brand “Wisdom of Krauts” which is still not transcripted in the desired way. The correct spelling seems to dominate the process and we get the undesired transcription result even after putting it into the list of special words. 😦
The only questfox way to “correct” this spelling would be the use of Regular Expressions in the questlogix. By the way: questlogix is such a distinct word with no similarity in the dictionary that our integration in the list of special words did function right away. Here is how to use RegEx in questfox:
Even while still in development we hope that you appreciate the publication of this feature and share your experience with us.
At first we defined the usage based on each questfox project as we believe that every specific project might have the same vocabulary issues inside one project. You can define your individual “Bag of Words” in the projects settings
The syntax is separate each word by comma. If you have several words just use them in a row
Word, word, word, combination of words,
Our typical list would be
questfox, pangea labs, questlogix, what’s your quest?, Speech2Text
In questfox you can paste one list of words per project under
-> Special List of Words for Transcription / Out of Vocabulary (OOV)
Please share your gained knowledge about this feature.
Voices of pangea: 120 languages supported in voice and video question types in questfox for real-time transcriptionPosted: 4. November 2019
pangea or pangaea?
The questfox development is managed by the Swiss company pangea labs. The idea of “pangea” was developed by the German researcher Alfred Wegener who described in this work “The Origin of Continents” (1912) the idea of a supercontinent, in times when all continents have been one.
Our company was called pangea labs after a project in 120 countries after which we reconsidered our business figuring that our mission was to bring all people on this now different planet back together again with software tools. This happened in 2006 in just one language which was English.
We are more than proud to announce a real pangea feature with the implementation of a set of 120 languages and variants of languages inside of questfox from now on. It is now possible to interview using the voice of the customer in the languages described underneath. questfox will be able to generate a text transcript from the selected language.
120 speech variants ready to be used in questfox
A user can select from a list of the following variants of a language. We know that this is a major step for the development of market research. We are aware of the fact that now everyone will be able to use that technology yet, but our mission is to support the planet with these kinds of tools to re-create the supercontinent on the basis of software.
The integrated languages are available for Voice and even video question types. The text-to-speech reading features are not yet supported.
Here is the list of languages already available in questfox (November 2019):
|Language||Language (German Name)|
|Afrikaans (Suid-Afrika)||Afrikaans (Südafrika)|
|አማርኛ (ኢትዮጵያ)||Amharisch (Äthiopien)|
|Հայ (Հայաստան)||Armenisch (Armenien)|
|Azərbaycan (Azərbaycan)||Aserbaidschanisch (Aserbaidschan)|
|Bahasa Indonesia (Indonesia)||Indonesisch (Indonesien)|
|Bahasa Melayu (Malaysia)||Malaiisch (Malaysia)|
|বাংলা (বাংলাদেশ)||Bengalisch (Bangladesch)|
|বাংলা (ভারত)||Bengalisch (Indien)|
|Català (Espanya)||Katalanisch (Spanien)|
|Čeština (Česká republika)||Tschechisch (Tschechische Republik)|
|Dansk (Danmark)||Dänisch (Dänemark)|
|Deutsch (Deutschland)||Deutsch (Deutschland)|
|English (Australia)||Englisch (Australien)|
|English (Canada)||Englisch (Kanada)|
|English (Ghana)||Englisch (Ghana)|
|English (Great Britain)||Englisch (Großbritannien)|
|English (India)||Englisch (Indien)|
|English (Ireland)||Englisch (Irland)|
|English (Kenya)||Englisch (Kenia)|
|English (New Zealand)||Englisch (Neuseeland)|
|English (Nigeria)||Englisch (Nigeria)|
|English (Philippines)||Englisch (Philippinen)|
|English (Singapore)||Englisch (Singapur)|
|English (South Africa)||Englisch (Südafrika)|
|English (Tanzania)||Englisch (Tansania)|
|English (United States)||Englisch (USA)|
|Español (Argentina)||Spanisch (Argentinien)|
|Español (Bolivia)||Spanisch (Bolivien)|
|Español (Chile)||Spanisch (Chile)|
|Español (Colombia)||Spanisch (Kolumbien)|
|Español (Costa Rica)||Spanisch (Costa Rica)|
|Español (Ecuador)||Spanisch (Ecuador)|
|Español (El Salvador)||Spanisch (El Salvador)|
|Español (España)||Spanisch (Spanien)|
|Español (Estados Unidos)||Spanisch (USA)|
|Español (Guatemala)||Spanisch (Guatemala)|
|Español (Honduras)||Spanisch (Honduras)|
|Español (México)||Spanisch (Mexiko)|
|Español (Nicaragua)||Spanisch (Nicaragua)|
|Español (Panama)||Spanisch (Panama)|
|Español (Paraguay)||Spanisch (Paraguay)|
|Español (Perú)||Spanisch (Peru)|
|Español (Puerto Rico)||Spanisch (Puerto Rico)|
|Español (República Dominicana)||Spanisch (Dominikanische Republik)|
|Español (Uruguay)||Spanisch (Uruguay)|
|Español (Venezuela)||Spanisch (Venezuela)|
|Euskara (Espainia)||Baskisch (Spanien)|
|Filipino (Pilipinas)||Philippinisch (Philippinen)|
|Français (Canada)||Französisch (Kanada)|
|Français (France)||Französisch (Frankreich)|
|Galego (España)||Galizisch (Spanien)|
|ქართული (საქართველო)||Georgisch (Georgien)|
|ગુજરાતી (ભારત)||Gujarati (Indien)|
|Hrvatski (Hrvatska)||Kroatisch (Kroatien)|
|IsiZulu (Ningizimu Afrika)||Zulu (Südafrika)|
|Íslenska (Ísland)||Isländisch (Island)|
|Italiano (Italia)||Italienisch (Italien)|
|Jawa (Indonesia)||Javanisch (Indonesien)|
|ಕನ್ನಡ (ಭಾರತ)||Kannada (Indien)|
|ភាសាខ្មែរ (កម្ពុជា)||Khmer (Kambodscha)|
|ລາວ (ລາວ)||Lao (Laos)|
|Latviešu (latviešu)||Lettisch (Lettland)|
|Lietuvių (Lietuva)||Litauisch (Litauen)|
|Magyar (Magyarország)||Ungarisch (Ungarn)|
|മലയാളം (ഇന്ത്യ)||Malayalam (Indien)|
|मराठी (भारत)||Marathi (Indien)|
|Nederlands (Nederland)||Niederländisch (Niederlande)|
|नेपाली (नेपाल)||Nepalesisch (Nepal)|
|Norsk bokmål (Norge)||Norwegisch (Norwegen)|
|Polski (Polska)||Polnisch (Polen)|
|Português (Brasil)||Portugiesisch (Brasilien)|
|Português (Portugal)||Portugiesisch (Portugal)|
|Română (România)||Rumänisch (Rumänien)|
|සිංහල (ශ්රී ලංකාව)||Singhalesisch (Sri Lanka)|
|Slovenčina (Slovensko)||Slowakisch (Slowakei)|
|Slovenščina (Slovenija)||Slowenisch (Slowenien)|
|Urang (Indonesia)||Sundanesisch (Indonesien)|
|Swahili (Tanzania)||Swahili (Tansania)|
|Swahili (Kenya)||Swahili (Kenia)|
|Suomi (Suomi)||Finnisch (Finnland)|
|Svenska (Sverige)||Schwedisch (Schweden)|
|தமிழ் (இந்தியா)||Tamilisch (Indien)|
|தமிழ் (சிங்கப்பூர்)||Tamilisch (Singapur)|
|தமிழ் (இலங்கை)||Tamilisch (Sri Lanka)|
|தமிழ் (மலேசியா)||Tamilisch (Malaysia)|
|తెలుగు (భారతదేశం)||Telugu (Indien)|
|Tiếng Việt (Việt Nam)||Vietnamesisch (Vietnam)|
|Türkçe (Türkiye)||Türkisch (Türkei)|
|اردو (پاکستان)||Urdu (Pakistan)|
|اردو (بھارت)||Urdu (Indien)|
|Ελληνικά (Ελλάδα)||Griechisch (Griechenland)|
|Български (България)||Bulgarisch (Bulgarien)|
|Русский (Россия)||Russisch (Russland)|
|Српски (Србија)||Serbisch (Serbien)|
|Українська (Україна)||Ukrainisch (Ukraine)|
|עברית (ישראל)||Hebräisch (Israel)|
|العربية (إسرائيل)||Arabisch (Israel)|
|العربية (الأردن)||Arabisch (Jordanien)|
|العربية (الإمارات)||Arabisch (Vereinigte Arabische Emirate)|
|العربية (البحرين)||Arabisch (Bahrain)|
|العربية (الجزائر)||Arabisch (Algerien)|
|العربية (السعودية)||Arabisch (Saudi-Arabien)|
|العربية (العراق)||Arabisch (Irak)|
|العربية (الكويت)||Arabisch (Kuwait)|
|العربية (المغرب)||Arabisch (Marokko)|
|العربية (تونس)||Arabisch (Tunesien)|
|العربية (عُمان)||Arabisch (Oman)|
|العربية (فلسطين)||Arabisch (Palästina)|
|العربية (قطر)||Arabisch (Katar)|
|العربية (لبنان)||Arabisch (Libanon)|
|العربية (مصر)||Arabisch (Ägypten)|
|فارسی (ایران)||Persisch (Iran)|
|हिन्दी (भारत)||Hindi (Indien)|
|ไทย (ประเทศไทย)||Thailändisch (Thailand)|
|한국어 (대한민국)||Koreanisch (Südkorea)|
|國語 (台灣)||Chinesisch, Mandarin (traditionell, Taiwan)|
|廣東話 (香港)||Chinesisch, Kantonesisch (traditionell, Hongkong)|
|普通話 (香港)||Chinesisch, Mandarin (vereinfacht, Hongkong)|
|普通话 (中国大陆)||Chinesisch, Mandarin (vereinfacht, China)|
3 Clicks away from your mother tongue in research
The languages are selected under question settings
Using speech in research still requires some technical tests upfront, but is a fantastic new feature and broadens the opportunities of research. We will continuously work on these feature sets and support the world with more and more pangea ideas soon.
How to analyze speech?
The voice of a person is transcribed into written text in questfox. As an option you can save the spoken word as an audio file. At the moment questfox will only offer you a look at the data but is not providing any specific analytic tools for voice or text. Little by little we develop these analytic parts together with software partners specialized in that field of research. Over time more and more partners will be able to offer analytic tools in their specific language. That is the return of the supercontinent pangea.
We are very excited to announce the integration of an application programming interface (API) inside of questions. questfox is partnering with audEERING and their emotion analysis engine based on speech analytics. questfox is now offering a permanent link to the incredible innovative audio analytics tool from audEERING (Munich).
Some questfox users already have access to a new question type called Speech2Text Voice Analysis. The question type is available for some users under Multimedia Insights.
The question type is basically a speaking answer question type (with all the restrictions still in practice*), where the transcript is generated based on the audio file. On top of this functionality, the audiofile is sent via API to audEERING to the check for emotional patterns in the voice answer. Within seconds questfox is able to receive the answer in the form of numbers assigned to different emotional states of a person. These values are saved inside the question type and kept in the background for export or other purposes. For example they can be used right away used for different actions during the live Interview. The values are exported alongside other data in the export file.
At the moment we are still testing the performance of the integrated API inside a survey tool. After additional tests runs, this emotional tracking opportunity will allow totally different kinds of research.
This new function is not included in the standard questfox software package as it causes additional cost anytime it is used. After a defined testing time we will define the pricing for this fabulous new feature set. Little by little we will create templates of ideas how to use emotional tracking in the interview process. We are also looking forward to see our users fantastic ideas come to life with this new set of opportunities.
We are looking forward to see new forms of research with the integration of the cognitive API inside the survey tool questfox. There may be more cognitive services connected to questfox in the future. We strongly believe in the power of expert knowledge from different fields of expertise. We will continue developing new standards in the world of interviewing as our field of expertise. The new technology allows us to connect several best-in-class approaches in one tool. The innovative tool allowing this open world of connected intelligence is still questfox.
*Restrictions when using voice type questions inside of questfox.
Years ago we made the decision not to develop an app for questfox as this would restrict our users around the world. We still believe that the effort needed for users to download an install an app for market research hardly pays off. This is why we opted for an open approach based on internet browser technologies. The negative side of this is that even in 2019 it is difficult to get voice answers from a series of browsers who do not allow the usage of a microphone in a browser. The biggest obstacle here still is Apple who does not allow to use a microphone in a browser situation on their devices. We strongly recommend a technology funnel in your suvey filtering out the users of iOS and some outdated browsers like Internet Explorer. Good news: Even Microsoft got the message right and the new version of Microsoft Edge is able to work with microphone. We know that it will still take years before 100% of a population can be researched with this approach. For the time being we have to live with the 70% of users who can actually be interviewed with this technology.
An indicator whether this is possible can be found here https://caniuse.com/#search=mediastream
We are very proud of our most recent development in speech2text recognition and automated analysis for market research. While still testing the new technology we face issues with some browsers in the market. Our approach is “device agnostic” and should be functional independently from devices or operating systems.
A lot of decision makers tend to use Apple devices and struggle with the new Talk2Me functionalities on their very expensive phones. They right away believe that something is wrong with questfox. Sorry guys. We are ahead of times in terms of development compared to Apple when it comes to media capturing in browser. This time it is not us. It is Apple slowing down innovation.
For the first time in our development we have to aknowledge, that we cannot keep that device agnostic promise for our most recent development. Apparently our developers are ahead of the development of Apple with Safari. The new multimedia functions inside of questfox will only become usable on iOS and Safari with the launch of their new version 11. We are hoping for a quick release within the Apple environment, but this is definitively outside of our reach.
If you want to stay updated about the different functionalities check this website from time to time: http://caniuse.com/#search=getUserMedia()
P.S. Internet Explorer is officially dead, but still operating on millions of computers. Hopefully Apple is forcing updates into their user base. Anyhow we try to stay in the first row of innovation knowing that we will lose some people sticking to older technology.
Rensis Likert would love questfox. Examples of animated scale examples for market research in questfoxPosted: 2. June 2016
The Likert Scale is the origin of all rating scales. We have been using graphical scales for the last 10 years successfully. In the development of mobile research methods we see an even stronger need for graphical interaction.
The opportunities of questfox allow to use different scales with adapted sliders. The scale can be used with a mouse or with a touch event.
On mobile devices the scale is responsive, but the relative dimension of the scale stays the same.
To inspire you to create your own individualized scales we are sharing some animated examples of scales used with questfox.
Hoping to see more creative scales from the questfox community.
Thanks to Rensis Likert who is and always be the father of modern scaling. We guess he would appreciate some of our ideas.
questfox questionnaires could always look better than the average online questionnaire out there. From today on it is much easier to create individually designed social interactions with questfox.
The new “Project Style” definitions do no longer require additional skills to make your research approach look beautiful or individual. Without any knowledge of CSS you can individually design background images, Next&Back buttons, interaction buttons and much more.
An example with very colorful backgrounds can be seen here:
The features are available under CREATE SETTINGS – PROJECT STYLE
The new version of questfox contains a brand new export functionality which allows to use Excel sheets of unlimited size. The standard XLS export publishes three spreadsheets containing a short field report, all the data gathering with interviews and a variable description spreadsheet. The overall process now takes less time and offers you a very quick approach to your data. We will continue working on the design. In the future additional and personal design will be possible for professional users.
Users with access to the branding feature of questfox are now able to test a limited set of CSS definitions in order to re-define the questionnaire design.
The new opportunities of cascading stylesheets will continouasly be developed.
The idea is to centralize changes of graphical settings inside questfox to corporate design.
Please allow some more testing before all features can be used with a defined CSS.
We will let you know when all elements of questfox can be defined via CSS and Provider a manual for the manipulation of CSS.
The pangea labs team has been working on some usability features of questfox.
Even though such a development never comes to an end, we are quite satisfied with the results.
Any comments on how to further improve are welcome.
Hope that we made your days more efficient.