Repository Issue Activity (beta)

#320 Startup

#339 Related work: Fun-ASR-Nano — audio encoder + LLM for speech understanding

#338 Feature Request: Consider SenseVoice/Paraformer as ASR component option

#337 Feature: Consider SenseVoice encoder as alternative audio frontend

#317 AttributeError: 'Gemma3Config' object has no attribute 'num_hidden_layers'

#335 [Partnership] Scoped integration collaboration

#333 deploying this ultravox model hugginface on vllm

#332 📋 Documentation Enhancement Suggestion

#311 v0.6 inference code for conversation mode doesn't work for v0.5 models

#331 React Native SDK stuck in 'connecting' state - session.status never transitions

#318 Batch size during training

#324 Audio chat only works for first message in gradio_demo.py

#323 Service Speed vs Self-Hosted Speed

#322 Error while running ultravad - ValueError: Text contains too many audio placeholders. (Expected 1 placeholders)

#321 License of Benchmark (fixie-ai/spoken-web-questions, fixie-ai/trivia_qa-audio)

#314 Config for model with gemma 270m

#273 FYI: here's an issue for livekit's agents to demo ultravox

#316 Persian Language Support in Ultravox Realtime

#8 Evaluate Ultravox performance when quantized

#293 Not able to fixie-ai/ultravox-v0_5-llama-3_1-8b on 4090 24gb vram machine

#315 Required Audio Hours for Supporting a New Language

#307 0.6 source?

#313 TTS with Resemble AI

#308 `recordingEnabled` API flag not respected

#306 Feature: Increase ASR and language understanding for tagalog