Live data from Hacker News

Whisper – open source speech recognition by OpenAI

openai.com

281–290 of 508 posts

Re: Whisper – open source speech recognition by OpenAI

#281
post #85

Earlier quoted context omitted.

It was already better. I edit a podcast and have > a decade of pro audio editing experience in the film industry, and I was already using a commercial AI transcription service to render the content to text and sometimes edit it as such (outputting edited audio). Existing (and affordable) offerings are so good that they can cope with shitty recordings off a phone speaker and maintain ~97% accuracy over hour-long conve…

>~97% accuracy over hour-long conversations. I'm sure it's been an absolute godsend for law enforcement 97% accuracy means roughly three or four errors per minute of speech. That seems potentially extremely problematic for something like law enforcement use where decisions with significant impact on people's day and/or life might be made on the basis of "evidence".

I imagine a certain percentage of a given population is on a voice call at any one time.

1. Set up a computer with voice recognition software that flags certain patterns.

2. Connect computer to voice call communication network.

3. Configure computer to switch between calls every x number of seconds.

Think of it like a system to generate leads for law enforcement that can be integrated with other systems to produce the best quality leads.

Re: Whisper – open source speech recognition by OpenAI

#282
post #85

Earlier quoted context omitted.

It was already better. I edit a podcast and have > a decade of pro audio editing experience in the film industry, and I was already using a commercial AI transcription service to render the content to text and sometimes edit it as such (outputting edited audio). Existing (and affordable) offerings are so good that they can cope with shitty recordings off a phone speaker and maintain ~97% accuracy over hour-long conve…

>~97% accuracy over hour-long conversations. I'm sure it's been an absolute godsend for law enforcement 97% accuracy means roughly three or four errors per minute of speech. That seems potentially extremely problematic for something like law enforcement use where decisions with significant impact on people's day and/or life might be made on the basis of "evidence".

Not really. Imagine that they do simple keyword matching on the text. Anything that's missed (part of the 97%) the criminals get away with. Anything that matches in the 3% is then checked by a human (by listening to the audio at that time stamp). So you only need to manually check the 3%, and even then only if something you're interested in is found.

Re: Whisper – open source speech recognition by OpenAI

#284
post #251

Earlier quoted context omitted.

Was this with the default base model, or the medium or large model? This can be specified with the —model flag.

I meant the 'Google's recorder app' from the parent comment and not Whisper.

Ah right, sorry got my comment threads mixed up! Someone else was asking about performance with accented English speakers in another comment.

Re: Whisper – open source speech recognition by OpenAI

#286
post #118

Earlier quoted context omitted.

The company name could be 100% legit, there is nothing stopping you from a forming a company with that name and not even sell beef.

It definitely happens. There are at least two companies that have branded [..] Kosher Gelatin™. One of them makes gelatin that is considered non-kosher by all of the major kashrus agencies. "Kosher Gelatin®", when in the ingredients, just means the product contains pork.

For what it's worth, I've spent a few minutes googling and can't find any story that corroborates this. The only US trademark I can find around "kosher gelatin" is by the brand Kolatin, which is apparently certified Kosher.

Re: Whisper – open source speech recognition by OpenAI

#287
post #219

Here's a live demo on Hugging Face Spaces if you want to try - https://huggingface.co/spaces/Amrrs/openai-whisper-live-tran...

I've tried speaking to that demo several times... I used the built in feature to record from microphone, and I played back the samples to make sure they were audible and clear.

Sometimes it outputs the words "thank you" (which I did not say), sometimes it outputs a period. It never once output anything I said. It seems completely broken.

EDIT: apparently something about the combination of Safari+HF+Whisper was not working. I tried another Whisper demo on HF and had the same results. Switching to Chrome made it work flawlessly... I have no idea what kind of codec incompatibility was happening.

Re: Whisper – open source speech recognition by OpenAI

#288

Japanese results looks pretty impressive! Took マッコウクジラ14頭が海岸に打ち上げられる オーストラリア(2022年9月21日) https://www.youtube.com/watch?v=bZkNIzeRBk4 Extracted audio with youtube-dl -f bestaudio https://www.youtube.com/watch\?v\=bZkNIzeRBk4 Converted into [00:00.000 --> 00:13.000] オーストラリア南部の島で、真っ向くじら14棟が海岸に打ち上げられて死んでいるのが見つかり、専門家が調査のため原地入りしました。 [00:13.000 --> 00:25.000] 原地メディアによりますと、オーストラリア南部のキング棟で、19日、少なくとも14棟の真っ向くじらが海岸に打ち上げられて死んでいるの…

Did you try translating them to english? I want to see if you get a similar error as me with a random phrase "Translated by Releska" showing up.

It's called hallucination. As the model is trained on unsupervised data, such errors do seldom happen. The model picks up that such phrases occur in translations and inserts them even if they do not appear in the source. This is described in the paper.

Re: Whisper – open source speech recognition by OpenAI

#289

Their Scottish accent example is pretty good, I'd like to see it work on some very strong English accents like this one: https://www.youtube.com/watch?v=nJ7QB3om-QY

Are you sure? I just ran some of Kimmy's sketches through it and ... The results are garbage.

Re: Whisper – open source speech recognition by OpenAI

#290

Earlier quoted context omitted.

Their models range from 70mb to 3gb. The largest model is smaller than the optimised stable diffusion. Not sure what the inference speed is like, haven't tried it myself yet.

I just tested it myself. Its fast enough on colab, couple of seconds but not sure if its fast enough to transcribe realtime audio yet.

"small" runs in realtime on Macbook Air M1 CPU.
Post reply on HN