Tag: gender

How well do Google and Microsoft and recognize speech across dialect, gender and race?

~ Rachael Tatman ~ 1 Comment

If you’ve been following my blog for a while, you may remember that last year I found that YouTube’s automatic captions didn’t work as well for some dialects, or for women. The effects I found were pretty robust, but I wanted to replicate them for a couple of reasons:

  • I only looked at one system, YouTube’s automatic captions, and even that was over a period of several years instead of at just one point in time. I controlled for time-of-upload in my statistical models, but it wasn’t the fairest system evaluation.
  • I didn’t control for the audio quality, and since speech recognition is pretty sensitive to things like background noise and microphone quality, that could have had an effect.
  • The only demographic information I had was where someone was from. Given recent results that find that natural language processing tools don’t work as well for African American English, I was especially interested in looking at automatic speech recognition (ASR) accuracy for African American English speakers.

With that in mind, I did a second analysis on both YouTube’s automatic captions and Bing’s speech API (that’s the same tech that’s inside Microsoft’s Cortana, as far as I know).

Speech Data

For this project, I used speech data from the International Dialects of English Archive. It’s a collection of English speech from all over, originally collected to help actors sound more realistic.

I used speech data from four varieties: the South (speakers from Alabama), the Northern Cities (Michigan), California (California) and General American. “General American” is the sort of news-caster style of speech that a lot of people consider unaccented–even though it’s just as much an accent as any of the others! You can hear a sample here.

For each variety, I did an acoustic analysis to make sure that speakers I’d selected actually did use the variety I thought they should, and they all did.

Systems

For the YouTube captions, I just uploaded the speech files to YouTube as videos and then downloaded the subtitles. (I would have used the API instead, but when I was doing this analysis there was no Python Google Speech API, even though very thorough documentation had already been released.)

Bing’s speech API was a little  more complex. For this one, my co-author built a custom Android application that sent the files to the API & requested a long-form transcript back. For some reason, a lot of our sound files were returned as only partial transcriptions. My theory is that there is a running confidence function for the accuracy of the transcription, and once the overall confidence drops below a certain threshold, you get back whatever was transcribed up to there. I don’t know if that’s the case, though, since I don’t have access to their source code. Whatever the reason, the Bing transcriptions were less accurate overall than the YouTube transcriptions, even when we account for the fact that fewer words were returned.

Results

OK, now to the results. Let’s start with dialect area. As you might be able to tell from the graphs below, there were pretty big differences between the two systems we looked at. In general, there was more variation in the word error rate for Bing and overall the error rate tended to be a bit higher (although that could be due to the incomplete transcriptions we mentioned above). YouTube’s captions were generally more accurate and more consistent. That said, both systems had different error rates across dialects, with the lowest average error rates for General American English.

dialect
Differences in Word Error Rate (WER) by dialect were not robust enough to be significant for Bing (under a one way ANOVA) (F[3, 32] = 1.6, p = 0.21), but they were for YouTube’s automatic captions (F[3, 35] = 3.45,p < 0.05). Both systems had the lowest average WER for General American.
Now, let’s turn to gender. If you read my earlier work, you’ll know that I previously found that YouTube’s automatic captions were more accurate for men and less accurate for women. This time, with carefully recorded speech samples, I found no robust difference in accuracy by gender in either system. Which is great! In addition, the unreliable trends for each system pointed in opposite ways; Bing had a lower WER for male speakers, while YouTube had a lower WER for female speakers.

So why did I find an effect last time? My (untested) hypothesis is that there was a difference in the signal to noise ratio for male and female speakers in the user-uploaded files. Since women are (on average) smaller and thus (on average) slightly quieter when they speak, it’s possible that their speech was more easily masked by background noises, like fans or traffic. These files were all recorded in a quiet place, however, which may help to explain the lack of difference between genders.

gender
Neither Bing (F[1, 34] = 1.13, p = 0.29), nor YouTube’s automatic captions (F[1, 37] = 1.56, p = 0.22) had a significant difference in accuracy by gender.
Finally, what about race? For this part of the analysis, I excluded General American speakers, since they did not report their race. I also excluded the single Native American speaker. Even with fewer speakers, and thus reduced power, the differences between races were still robust enough to be significant for YouTube’s automatic captions and Bing followed the same trend. Both systems were most accurate for Caucasian speakers.

ethnicity
As with dialect, differences in WER between races were not significant for Bing (F[4, 31] = 1.21, p = 0.36), but were significant for YouTube’s automatic captions (F[4, 34] = 2.86,p< 0.05). Both systems were most accurate for Caucasian speakers.
While I was happy to find no difference in performance by gender, the fact that both systems made more errors on non-Caucasian and non-General-American speaking talkers is deeply concerning. Regional varieties of American English and African American English are both consistent and well-documented. There is nothing intrinsic to these varieties that make them less easy to recognize. The fact that they are recognized with more errors is most likely due to bias in the training data. (In fact, Mozilla is currently collecting diverse speech samples for an open corpus of training data–you can help them out yourself.)

So what? Why does word error rate matter?

There are two things I’m really worried about with these types of speech recognition errors. The first is higher error rates seem to overwhelmingly affect already-disadvantaged groups. In the US, strong regional dialects tend to be associated with speakers who aren’t as wealthy, and there is a long and continuing history of racial discrimination in the United States.

Given this, the second thing I’m worried about is the fact that these voice recognition systems are being incorporated into other applications that have a real impact on people’s lives.

Every automatic speech recognition system makes errors. I don’t think that’s going to change (certainly not in my lifetime). But I do think we can get to the point where those error don’t disproportionately affect already-marginalized people. And if we keep using automatic speech recognition into high-stakes situations it’s vital that we get to that point quickly and, in the meantime, stay aware of these biases.

If you’re interested in the long version, you can check out the published paper here.

Advertisement

Are “a female” and “a male” used differently?

~ Rachael Tatman ~ Leave a comment

In this first part of this two-post series, I looked at how “a male” and “a female” were used on Twitter. I found that one part of speech tagger tagged “male” as a proper noun really frequently (which is weird, cause it isn’t one) and that overall the phrase “a female” was waaaay more frequent. Which is  interesting in itself, since my initial question was “are these terms used differently?” and these findings suggest that they are. But the second question is how are these terms used differently? To answer this, we’ll need to get a little more qualitative with it.

Anas platyrhynchos male female
“Male” and “female” are fine for ducks, but a little weird for humans.
Using the same set of tweets that I collected last time, I randomly selected 100 tweets each from the “a male” and “a female” dataset. Then I hand tagged each subset of tweets for two things: the topic of the tweet (who or what was being referred to as “male” or “female”) and the part of speech of “male”  or “female”.

Who or what is being called “male” or “female”?

Rplot

Because there were so few tweets to analyze, I could do a content analysis. This is a methodology that is really useful when you don’t know for sure ahead of time what types of categories you’re going to see in your data. It’s like clustering that a human does.

Going into this analysis, I thought that there might be a difference between these datasets in terms of how often each term was used to refer to an animal, so I tagged tweets for that. But as I went through the tweets, I was floored by the really high number of tweets talking about trans people, especially Mack Beggs, a trans man from Texas who was forced to wrestle in the women’s division. Trans men were referred to as “a male” really, really often. While there wasn’t a reliable difference between how often “a female” and “a male” was used to refer to animals or humans, there was a huge difference in terms of how often they were  used to refer to trans people. “A male” was significantly more likely to be used to describe a trans person than “a female” (X2 (2, N = 200) = 55.33, p <.001.)

Part of Speech

Since the part of speech taggers I used for the first half of my analysis gave me really mixed results, I also hand tagged the part of speech of “male” or “female” in my samples. In line with my predictions during data collection, the only parts of speech I saw were nouns and adjectives.

When I looked at just the difference between nouns and adjectives, there was a little difference, but nothing dramatic. Then, I decided to break it down a little further. Rather than just looking at the differences in part of speech between “male” and “female”, I looked at the differences in part of speech and whether the tweet was about a trans person or a cis (not trans) person.

Rplot01
For tweets with “female”, it was used as a noun and an adjective at pretty much the same rates regardless of whether someone was talking about a trans person or a cis (non-trans) person. For tweets with “male”, though, when the tweet was about a trans person, it was used almost exclusively as a noun.

And there was a huge difference there. A large majority of tweets with “a male” and talking about a trans person used “male” as a noun. In fact, more than a third of my subsample of tweets using “a male” were using it as a noun to talk about someone who was trans.

So what’s going on here? This construction (using “male” or “female” as a noun to refer to a human) is used more often to talk about:

  1. Women. (Remember that in the first blog post looking at this, I found that “a female” is twice a common as “a male.)
  2. Trans men.

These both make sense if you consider the cultural tendency to think about cis men as, in some sense, the “default”. (Deborah Tannen has a really good discussion of this her article “Marked Women, Unmarked Men“. “Marked” is a linguistics term which gets used in a lot of ways, but generally means something like “not the default” or “the weird one”.) So people seem to be more likely to talk about a person being “a male” or “a female” when they’re talking about anyone but a cis man.

A note on African American English

giphy.gif

I should note that many of the tweets in my sample were African American English, which is not surprising given the large Black community on Twitter, and that use of “female” as a noun is a feature of this variety.  However, the parallel term used to refer to men in this variety is not “a man” or even “a male”, but rather “nigga”, with that spelling. This is similar to “dude” or “guy”: a nonspecific term for any man, regardless of race, as discussed at length by Rachel Jeantal here. You can see an example of this usage in speech above (as seen in the Netflix show “The Unbreakable Kimmy Schmidt“) or in this vine. (I will note, however, that it only has this connotation if used by a speaker of African American English. Borrowing it into another variety, especially if the speaker is white, will change the meaning.)

Now, I’m not a native user of African American English, so I don’t have strong intuitions about the connotation of this usage. Taylor Amari Little (who you may know from her TEDx talk on Revolutionary Self-Produced Justice) is, though, and tweeted this (quoted with permission):

If they call women “females” 24/7, leave em alone chile, run away

And this does square with my own intuitions: there’s something slightly sinister about someone who refers to women exclusively as “females”. As journalist Vonny Moyes pointed out in her recent coverage of ads offering women free rent in exchange for sexual favors, they almost refer to women as “girls or females – rarely ever women“. Personally, I find that very good motivation not to use “a male” or “a female” to talk about any human.

What’s up with calling a woman “a female”? A look at the parts of speech of “male” and “female” on Twitter .

~ Rachael Tatman ~ 5 Comments

This is something I’ve written about before, but I’ve recently had several discussions with people who say they don’t find it odd to refer to a women as a female. Personally, I don’t like being called “a female” becuase its a term I to associate strongly with talking about animals. (Plus, it makes you sound like a Ferengi.)  I would also protest men being called males, for the same reason, but my intuition is that that doesn’t happen as often. I’m willing to admit that my intuition may be wrong in this case, though, so I’ve decided to take a more data-driven approach. I had two main questions:

  • Do “male” and “female” get used as nouns at different rates?
  • Does one of these terms get used more often?

Data collection

I used the Twitter public API to collect two thousand English tweets, one thousand each containing the exact string “a male” and “a female”. I looked for these strings to help get as many tweets as possible with “male” or “female” used as a noun. “A” is what linguist call a determiner, and a determiner has to have a noun after it. It doesn’t have to be the very next word, though; you can get an adjective first, like so:

  • A female mathematician proved the theorm.
  • A female proved the theorm.

So this will let me directly compare these words in a situation where we should only be able to see a limited number of possible parts of speech & see if they differ from each other. Rather than tagging two thousand tweets by hand, I used a Twitter specific part-of-speech tagger to tag each set of tweets.

A part of speech tagger is a tool that guesses the part of speech of every word in a text. So if you tag a sentence like “Apples are tasty”, you should get back that “apples” is a plural noun, “are” is a verb and “tasty” is an adjective. You can try one out for yourself on-line here.

Parts of Speech

In line with my predictions, every instance of “male” or “female” was tagged as either a noun, an adjective or a hashtag. (I went through and looked at the hashtags and they were all porn bots. #gross #hazardsOfTwitterData)

However, not every noun was tagged as the same type of noun. I saw three types of tags in my data: NN (regular old noun), NNS (plural noun) and, unexpectedly, NNP (proper noun, singular). (If you’re confused by the weird upper case abbreviations, they’re the tags used in the Penn Treebank, and you can see the full list here.) In case it’s been a while since you studied parts of speech, proper nouns are things like personal or place names. The stuff that tend to get capitalized in English. The examples from the Penn Treebank documentation include “Motown”, “Venneboerger”,  and “Czestochwa”. I wouldn’t consider either “female” or “male” a name, so it’s super weird that they’re getting tagged as proper nouns. What’s even weirder? It’s pretty much only “male” that’s getting tagged as a proper noun, as you can see below:

maleVsFemalePOS
Number of times each word tagged as each part of speech by the GATE Twitter part-of-speech tagger. NNS is a plural noun, NNP a proper noun, NN a noun and JJ an adjective.

The differences in tagged POS between “male” and “female” was super robust(X2(6, N = 2033) = 1019.2, p <.01.). So what’s happening here?  My first thought was that it might be that, for some reason, “male” is getting capitalized more often and that was confusing the tagger. But when I looked into, there wasn’t a strong difference between the capitalization of “male” and “female”: both were capitalized about 3% of the time. 

My second thought was that it was a weirdness showing up becuase I used a tagger designed for Twitter data. Twitter is notoriously “messy” (in the sense that it can be hard for computers to deal with) so it wouldn’t be surprising if tagging “male” as a proper noun is the result of the tagger being trained on Twitter data. So, to check that, I re-tagged the same data using the Stanford POS tagger. And, sure enough, the weird thing where “male” is overwhelming tagged as a proper noun disappeared.

stanfordTaggerPOS
Number of times each word tagged as each part of speech by the Stanford POS tagger. NNS is a plural noun, NNP a proper noun, NN a noun, JJ an adjective and FW a “foreign word”.

So it looks like “male” being tagged as a proper noun is an artifact of the tagger being trained on Twitter data, and once we use a tagger trained on a different set of texts (in this case the Wall Street Journal) there wasn’t a strong difference in what POS “male” and “female” were tagged as.

Rate of Use

That said, there was a strong difference between “a female” and “a male”: how often they get used. In order to get one thousand tweets with the exact string “a female”, Twitter had to go back an hour and thirty-four minutes. In order to get a thousand tweets with “a male”, however, Twitter had to go back two hours and fifty eight minutes. Based on this sample, “a female” gets said almost twice as often as “a male”.

So what’s the deal?

  • Do “male” and “female” get used as nouns at different rates?  It depends on what tagger you use! In all seriousness, though, I’m not prepared to claim this based on the dataset I’ve collected.
  • Does one of these terms get used more often? Yes! Based on my sample, Twitter users use “a female” about twice as often as “a male”.

I think the greater rate of use of “a female” that points to the possibility of an interesting underlying difference in how “male” and “female” are used, one that calls for a closer qualitative analysis. Does one term get used to describe animals more often than the other? What sort of topics are people talking about when they say “a male” and “a female”? These questions, however, will have to wait for the next blog post!

In the meantime, I’m interested in getting more opinions on this. How do you feel about using “a male” and “a female” as nouns to talk about humans? Do they sound OK or strike you as odd?

My code and is available on my GitHub.

Preference for wake words varies by user gender

~ Rachael Tatman ~ 2 Comments

I recently read a very interesting article on the design of aspects of choosing a wake word, the word you use to turn on a voice-activated system. In Star Trek it’s “Computer”, but these days two of the more popular ones are “Alexa” and “OK Google”. The article’s author was a designer and noted that she found “Ok Google” or “Hey Google” to be more pleasant to use than “Alexa”. As I was reading the comments (I know, I know) I noticed that a lot of the people who strongly protested that they preferred “Alexa” had usernames or avatars that I would associate with male users. It struck me that there might be an underlying social pattern here.

So, being the type of nerd I am, I whipped up a quick little survey to look at the interaction between user gender and their preference for wake words. The survey only had two questions:

  • What is your gender?
    • Male
    • Female
    • Other
  • If Google Home and the Echo offered identical performance in all ways except for the wake word (the word or phrase you use to wake the device and begin talking to it), which wake word would you prefer?
    • “Ok Google” or “Hey Google”
    • “Alexa”

I included only those options becuase those are the defaults–I am aware you can choose to change the Echo’s wake word. (And probably should, given recent events.) 67 people responded to my survey. (If you were one of them, thanks!)

So what were the results? They were actually pretty strongly in line with my initial observations: as a group, only men preferred “Alexa” to “Ok Google”. Furthermore, this preference was far weaker than people of other genders’ for “Ok Google”. Women preferred “Ok Google” at a rate of almost two-to-one, and no people of other genders preferred “Alexa”.

I did have a bit of a skewed sample, with more women than men and people of other genders, but the differences between genders were robust enough to be statistically significant (c2(2, N = 67) = 7.25, p = 0.02)).

genderandwakewords
Women preferred “Ok Google” to “Alexa” 27:11, men preferred “Alexa” to “Ok Google” 14:11, and the four people of other genders in my survey all preferred “Ok Google”.

So what’s the take-away? Well, for one, Johna Paolino (the author of the original article) is by no means alone in her preference for a non-gendered wake word. More broadly, I think that, like the Clippy debacle, this is excellent evidence that there are strong gendered differences in how users’ gender affects their interaction with virtual agents. If you’re working to create virtual agents, it’s important to consider all types of users or you might end up creating something that rubs more than half of your potential customers the wrong way.

My code and data are available here.

Google’s speech recognition has a gender bias

~ Rachael Tatman ~ 15 Comments

Edit, July 2020: Hello! This blog post has been cited quite a bit recently so I thought I’d update it with the more recent reserach. I’m no longer working actively on this topic, but in the last paper I wrote on it, in 2017, I found that when audio quality was controlled the gender effects disappeared. I take this to be evidence that differences in gender are due to differences in overall signal-to-noise ratio when recording in noisy environments rather than problems in the underlying ML models.

That said, bias against specific demographics categories in automatic speech recognition is a problem. In my 2017 study, I found that multiple commercial ASR systems had higher error rates for non-white speakers. More recent research has found the same effect: ASR systems make more errors for Black speakers than white speakers. In my professional opinion, the racial differences are both more important and difficult to solve.

The original, unedited blog post, continues below.

_____

In my last post, I looked at how Google’s automatic speech recognition worked with different dialects. To get this data, I hand-checked annotations  more than 1500 words from fifty different accent tag videos .

Now, because I’m a sociolinguist and I know that it’s important to stratify your samples, I made sure I had an equal number of male and female speakers for each dialect. And when I compared performance on male and female talkers, I found something deeply disturbing: YouTube’s auto captions consistently performed better on male voices than female voice (t(47) = -2.7, p < 0.01.) . (You can see my data and analysis here.)

accuarcyByGender
On average, for each female speaker less than half (47%) her words were captioned correctly. The average male speaker, on the other hand, was captioned correctly 60% of the time.

It’s not that there’s a consistent but small effect size, either, 13% is a pretty big effect. The Cohen’s d was 0.7 which means, in non-math-speak, that if you pick a random man and random woman from my sample, there’s an almost 70% chance the transcriptions will be more accurate for the man. That’s pretty striking.

What it is not, unfortunately, is shocking. There’s a long history of speech recognition technology performing better for men than women:

This is a real problem with real impacts on people’s lives. Sure, a few incorrect Youtube captions aren’t a matter of life and death. But some of these applications have a lot higher stakes. Take the medical dictation software study. The fact that men enjoy better performance than women with these technologies means that it’s harder for women to do their jobs. Even if it only takes a second to correct an error, those seconds add up over the days and weeks to a major time sink, time your male colleagues aren’t wasting messing with technology. And that’s not even touching on the safety implications of voice recognition in cars.

So where is this imbalance coming from? First, let me make one thing clear: the problem is not with how women talk. The suggestion that, for example, “women could be taught to speak louder, and direct their voices towards the microphone” is ridiculous. In fact, women use speech strategies that should make it easier for voice recognition technology to work on women’s voices.  Women tend to be more intelligible (for people without high-frequency hearing loss), and to talk slightly more slowly. In general, women also favor more standard forms and make less use of stigmatized variants. Women’s vowels, in particular, lend themselves to classification: women produce longer vowels which are more distinct from each other than men’s are. (Edit 7/28/2016: I have since found two papers by Sharon Goldwater, Dan Jurafsky and Christopher D. Manning where they found better performance for women than men–due to the above factors and different rates of filler words like “um” and “uh”.) One thing that may be making a difference is that women also tend not to be as loud, partly as a function of just being smaller, and cepstrals (the fancy math thing what’s under the hood of most automatic voice recognition) are sensitive to differences in intensity. This all doesn’t mean that women’s voices are more difficult; I’ve trained classifiers on speech data from women and they worked just fine, thank you very much. What it does mean is that women’s voices are different from men’s voices, though, so a system designed around men’s voices just won’t work as well for women’s.

Which leads right into where I think this bias is coming from: unbalanced training sets. Like car crash dummies, voice recognition systems were designed for (and largely by) men. Over two thirds of the authors in the  Association for Computational Linguistics Anthology Network are male, for example. Which is not to say that there aren’t truly excellent female researchers working in speech technology (Mari Ostendorf and Gina-Anne Levow here at the UW and Karen Livescu at TTI-Chicago spring immediately to mind) but they’re outnumbered. And that unbalance seems to extend to the training sets, the annotated speech that’s used to teach automatic speech recognition systems what things should sound like. Voxforge, for example, is a popular open source speech dataset that “suffers from major gender and per speaker duration imbalances.” I had to get that info from another paper, since Voxforge doesn’t have speaker demographics available on their website. And it’s not the only popular corpus that doesn’t include speaker demographics: neither does the AMI meeting corpus, nor the Numbers corpus.  And when I could find the numbers, they weren’t balanced for gender. TIMIT, which is the single most popular speech corpus in the Linguistic Data Consortium, is just over 69% male. I don’t know what speech database the Google speech recognizer is trained on, but based on the speech recognition rates by gender I’m willing to bet that it’s not balanced for gender either.

Why does this matter? It matters because there are systematic differences between men’s and women’s speech. (I’m not going to touch on the speech of other genders here, since that’s a very young research area. If you’re interested, the Journal of Language and Sexuality is a good jumping-off point.) And machine learning works by making computers really good at dealing with things they’ve already seen a lot of. If they get a lot of speech from men, they’ll be really good at identifying speech from men. If they don’t get a lot of speech from women, they won’t be that good at identifying speech from women. And it looks like that’s the case. Based on my data from fifty different speakers, Google’s speech recognition (which, if you remember, is probably the best-performing proprietary automatic speech recognition system on the market) just doesn’t work as well for women as it does for men.