I’m trying build with MARS8 TTS, rn available on Google Cloud and other compute platforms. As far as I can tell, they are the only Text to Speech on GCP’s Vertex AI platform. I see the new addition on Pipecat as well.
Supports 30-40 top languages, can run on any GCP/AWS location (probably great for network latency) and you get the model to run on your own Gpu, so no per token/character pricing. It’s by a company named Camb ai.
I’m trying build with MARS8 TTS, rn available on Google Cloud and other compute platforms. As far as I can tell, they are the only Text to Speech on GCP’s Vertex AI platform. I see the new addition on Pipecat as well.
Supports 30-40 top languages, can run on any GCP/AWS location (probably great for network latency) and you get the model to run on your own Gpu, so no per token/character pricing. It’s by a company named Camb ai.