Ajay Raj
Ajay Raj
Love the project, wondering if you'd like to use vocode's interfaces for speech synthesis! - ports the existing ElevenLabs, gTTS, and StreamElements integrations - enables Azure voices: https://azure.microsoft.com/en-us/products/cognitive-services/text-to-speech/ pros of...
[Vocode](https://github.com/vocodedev/vocode-python) provides a unified interface to many different TTS providers - this PR allows Langchain interactions to speak via a CallbackHandler. Adds `VocodeCallbackHandler` with two helper methods: `speak` and `speak_in_thread`:...
Rewrite input streaming to be built more bespoke into `StreamingConversation`: - Convert `BaseSynthesizer` into a worker that accepts - Convert the process of collating tokens into sentences into a worker...
removes `TypedModel` and all that comes with it and replaces it with `AdaptiveObject` allows us to properly move to pydantic v2 and not in half state as before, as well...
Different types of `AudioPipeline`s and various `*PhoneConversation`s have actions that only work if they are attached specifically to those pipelines in question - we need to manually create `ActionsWorker` and...
We’d like support other “streaming” conversations on the phone - basically anything that can accept streaming audio chunks and send out streaming audio chunks concurrently should supportable on a phone...
We use `TypedModel` to allow us to deserialize Configs to typed objects with discriminated unions when they are coming out of the ConfigManager, for example. It isn't great for a...
Once https://github.com/vocodedev/vocode-core/issues/651 is finished, release new contribution / usage docs for how to create new synthesizers and build new agents in the new abstractions
this PR: - decouples ActionsWorker from StreamingConversation - creates the notion of PipelineFactory and PipelineConfig, to be used with outbound calls / inbound call configs - removes coupling of PhoneConversation...