vosk-api VS whisper

So I don't know if my issue comes from my lack of knowledge of discord.js/voice or VOSK. so I guess the most important thing I need to see is if I am creating a proper stream for the Vosk API to capture the audio. if I can figure out how to capture an audio stream I can probably import that in to vosk and figure out how to use vosk myself. but right now I can't even get close! Thank you in advance...Sorry if this isn't the right place for this

I remember a while ago checking out the issues with Vosk speech recognition (written in C). A handful of it's issues are related to segfaults and null pointers.

first, good initiative! thanks for sharing. i think you gotta be more diligent and careful with the problem statement.

checking the weather in Sofia, Bulgaria requires cloud, current information. it's not "random speech". ESP SR capability issues don't mean that you cannot process it locally.

the comment was on "voice processing" i.e. sending speech to the cloud, not sending a call request to get the weather information.

besides, local intent detection, beyond 400 commands, there are great local STT options, working better than most cloud STTs for "random speech"

https://github.com/alphacep/vosk-api

I did a one-off text to speech tool for someone last year and had pretty good results with VOSK. One upside is that it works offline, although I imagine if you use TTS a lot you'll notice issues I didn't.

You can use vosk-api (https://github.com/alphacep/vosk-api) to listen to your audio, transform it to text, and then post the text to GPT-3, then using the vector sdk, have your responses said by vector.

The set up script wants to download https://github.com/alphacep/vosk-api/releases/download/v0.3.45/vosk-model-en-v0.3.45.zip, but this resource is not found. AFAICT all releases never contained a model file. Remedy: hardcode one model from https://alphacephei.com/vosk/models. I guessed and picked the one with the closest name, vosk-model-en-us-0.22.zip, just so I could continue.

Let’s not talk about local models as the hardware requirements are way beyond most of these people’s reach. I have a MacBook Air with an M2 chip and 8GB of RAM and can hardly run Whisper locally, so I use this HuggingFace space.

Last week, I launched notesGPT, a free and open source voice note app that has 35,000 visitors, 7,000 users, and over 1,000 GitHub stars so far in the last week. It allows you to record a voice note, transcribes it uses Whisper, and uses Mixtral via Together to extract action items and display them in an action items view. It’s also fully open source and comes equipped with authentication, storage, vector search, action items, and is fully responsive on mobile for ease of use.

I found a pretty good discussion in the topic here:

https://github.com/openai/whisper/discussions/1243

There is a plot of language performance on their repo: https://github.com/openai/whisper

I am not aware of a multi-lingual leaderboard for speech recognition models.

I don't know how OP does it, but here's how I'd do it:

* Generate a transcript by runing Whisper against the podcast audio file: https://github.com/openai/whisper

* Upload transcript to ChatGPT and ask it to summarize.

* Automate all the above.

Ahh, that makes sense. I've been building something like that, but only from other languages into English using Whisper

Whisper already generates subtitles[0], supporting VTT and SRT so this is just a thin wrapper around that.

[0]: https://github.com/openai/whisper/blob/e58f28804528831904c3b...

> although it does require you to wear headphones so the bot doesn't hear itself and get interrupted.

Maybe you can rely on some sort of speaker identification to sort this out?

https://github.com/openai/whisper/discussions/264

v3 only comes in one flavor: large.

I don’t think you’re going to have a good time running the large model on a Pi of any kind.

The large models are 32x slower than the tiny models, roughly.[0]

I’m seeing people report that the Pi 4 can transcribe 30 seconds of audio in somewhere between 30 seconds and 60 seconds with the tiny model.

You can do the math… 32x = 16 minutes to 32 minutes to transcribe 30 seconds of audio with the large model. Not a good time for most people.

The Pi 5 could be 2x to 3x faster.

I should benchmark my Pi 4 sometime (or Pi 5, if it ever shows up).

[0]: https://github.com/openai/whisper/blob/main/README.md#availa...