OpenAI Whisper

“I had amazing success with Spleeter, Demucs, MVSEP-MDX23-music-separation, DeepFilterNet, Whisper”

“competitors which have their own proprietary models or even use something like Pyannote nd whisper are certainly far better”

“The best workaround is a third party app that lets you run Parakeet V3 or Whisper on your phone.”

“Still hanging on to Whisper for transcribing and translating audio in real time locally.”

“this uses the Whisper Large model which is quite a bit better than the Youtube auto generated ones”

“It's really annoying especially because Whisper is really quite good otherwise.”

“tends to generate plausible-in-context content when there are long gaps or silence”

“I can almost always live with the few mistakes because the overall stream-of-thought word-salad I provide is still clear”

“Combined with Whisper+ from F-Droid and you have a nice way to input voice as text from everywhere.”

“the open whisper model which really is great especially if you use the larger parameter versions”

“if it captures at least a few bits of interesting information that would otherwise go unnoticed it's still a win”

“I have built an ambient todo list / health data extractor using Gemma 4 2EB and Whisper”

“I read somewhere it comes from training on YouTube audio and happens when there is silence”

“Parakeet and Whisper are much smaller and perform great for quick dictation and transcription of longer files”

“The latest open source local STT models people are running on devices are significantly more robust”

“Ever since whisper came out, usefulness for AST models running locally suddenly became a thing.”

“One thing all these models do (including whisper) is just omit full sentences, it's the worst thing a model can do.”

“they hallucinate when they misunderstand and tend to come up with random phrases”

“Whisper often mistook actual words I say and hallucinate all the time when speaking technical jargon.”

“Whisper is actually better than any other model that I've seen, even things like Parakeet.”

“I already have run all my downloaded podcasts through whisper to get transcripts with timestamps”

“My ~1.7% WER and faster than realtime processing in my application make it more than adequate.”

“We can fit 4 versions of the model in memory at once on a ~$1/hr A10 and have half the VRAM leftover.”

“When OpenAI released Whisper, it was by far best-in-class, and they haven't released any major upgrades since then.”

“free, OSS dictation tools that can run on-device, that are so much better than Apple's dictation”

“Whisper remains the gold standard - you can run it locally with good accuracy across languages”

“The problem with Whisper is not the inference quality: medium and large are incredible.”

“This combo has almost unbeatable accuracy and it rejects noises in the background really well.”

“Whisper is really good for 1 person speaking. With more people you get repetitions.”

“A local pipeline with Whisper and Llama 3 is viable now and solves the privacy issue.”

“I always get much better results in a lot faster time when i just turn on the mic and have whisper transcribe”

“The issue with Whisper is that it messes up both names of both products and guests.”

“having real trouble to set up working environment for any Whisper derivatives after migrating to Fedora”

“all of the state-of-the-art speech-to-text systems are large-language model based and are significantly faster and better than what we had before”

“polishing Whisper speech to text output, so I can dictate things to my computer and get coherent sentences”

“generated subtitles with Whisper (with every single sentence having at least some mistake)”

“current versions of Whisper GPU will not run on my not-quite-10-year old Pascal GPU anymore”

“For this use case I think best bet is still a toolchain with a transcription model like whisper fed into an LLM to summarize”

“Installing and running Whisper locally just to get notifications seems like more resource consumption than some of us are looking for.”

“It might make up a bunch of words, like "subtitles by soandso", when there's silence”