Does anyone know what the cheapest service (that works well) / API that allows feeding videos / audio and getting transcripts of the audio? Preferably something that can manage the cocktail party problem somewhat efficiently?
Google speech to text, etc.?
Conversation
Replying to
Not sure about APIs. At scale most of them become quite expensive. The cheapest option I know of is kaldi, it's really inexpensive at scale - developer.nvidia.com/blog/nvidia-ac
1
Replying to
I've used this previously for a one-off. It worked well for the job, but might be a bit pricey for what you need. happyscribe.com