Amazon AWS,  Technology

Speech to Text Accuracy

I wanted to test the relative accuracy of various speech to text engines. I tested Line 1, Amazon, Deepgram and Google. All vastly exceeded my expectations.

Here is my original text. I tried doing a normal speaking voice. I left myself a voicemail and used the resulting recording as the basis for the other tests, so all tests were from the same file:

"The quick brown fox jumps over the lazy dog.

Fourscore and seven years ago our fathers brought forth, on this continent, a new nation, conceived in liberty, and dedicated to the proposition that all men are created equal. 

His palms are sweaty, knees weak, arms are heavy
There's vomit on his sweater already, mom's spaghetti
He's nervous, but on the surface he looks calm and ready
To drop bombs, but he keeps on forgettin'
What he wrote down, the whole crowd goes so loud
He opens his mouth, but the words won't come out
He's chokin', how, everybody's jokin' now
The clocks run out, times, over, blaow"

I was impressed. All four services did an impressive job. More than anything it comes down to formatting more than context. For all of them you absolutely get the context.

Voicemail – Human Transcription – Line 1 Communications
The Quick Brown Fox Jumps over the lazy dog. Four score and 7 years ago our father brought forth on this continent a new nation conceived in liberty and delegated to the proposition that all men are created equal. His palms are sweaty knees weak arms are heaving there is vomit on his sweater already, moms spaghetti. Hes nervous but on the surface he looks calm and ready. To drop bombs but he keeps on forgetting what he wrote down the whole crowd grows so loud. He’s choking, how? Everyone’s joking now. the clocks run out times over. Blow.

Line 1: Overall not bad – Formatting is good and context is great. Only two words incorrect.

Amazon
The quick brown fox jumps over the lazy dog four score, And seven years ago, our fathers brought forth on this continent, a new nation conceived in liberty and dedicated to the proposition that all men are created equal.
His palms are sweaty knees. Weak arms are heavy. There’s vomit on his sweater. Already Mom’s spaghetti. He’s nervous, but on the surface, he looks calm and ready to drop bombs, But he keeps on forgetting.
What he wrote down. The whole crowd goes so loud. He opens his mouth, but the words won’t come out. He’s choking how everyone’s joking. Now the clocks run out. Time’s over.
Wow

Amazon: Again, great context, but the formatting could be much better.

Deepgram
The quick Brown Fox jumps over the lazy dog four score and seven years ago our fathers brought forth on this continent a new nation conceived in liberty and dedicated to the proposition that all men are created equal his palms are sweaty knees weak arms are heavy there’s vomit on his sweater already mom’s spaghetti he’s nervous but on the surface.

He looks calm and ready to drop bombs, but it keeps on forgetting when he wrote down the whole crowd grows so loud he opens his mouth but the words won’t come out he’s choking how everyone’s joking now the clocks run out times over wow.

Deepgram: Better formatting and no real missed words.

Google
The quick brown fox jumps over the lazy dog xx four score and seven years ago. Our fathers brought forth on this continent a new nation conceived in liberty and dedicated to the proposition that all men are created equal.
His palms are sweaty knees weak arms are heavy. There’s vomit on his sweater already mom’s spaghetti. He’s nervous, but on the surface, he looks calm and ready to drop bombs, but he keeps on forgetting what he wrote down. The whole crowd goes so loud. He opens his mouth, but the words won’t come out. He’s choking how everyone joke and now the clocks run out times over. Wow.

Google: A little surprised at the missed words at the end and no real formatting.

Leave a Reply

Your email address will not be published. Required fields are marked *