Topics / #speech-to-speech
Speech-to-speech projects on GitHub
As of , the most-starred Speech-to-speech projects on GitHub are personaplex (12.1K stars), speech-swift (1.3K stars) and dsh-voice-ai-girlfriend (1.1K stars). GitHub.Locker tracks 5 Speech-to-speech repositories with 15.6K stars combined. The fastest-growing on November 17 was sophon-realtime, with +15 stars in a day.
Hot right now
Most stars gained in a day · Nov 17
+15sophon-realtime
adventists-ai/sophon-realtime · ★ 436An open-source family of full-duplex voice models: it listens while you talk, hears your tone and answers with a voice that can laugh, sigh and breathe.
+11ai-native-callcenter
rasonyang/ai-native-callcenter · ★ 677An open-source, self-hosted AI call center: calls land on FreeSWITCH and are answered by a speech-to-speech voice model by default, with a flow deciding when to hand the caller to a real human agent in a queue. It ships as one Go binary with a REST+SSE API, OpenAPI spec, PostgreSQL and FreeSWITCH underneath.
+4personaplex
NVIDIA/personaplex · ★ 12.1KPersonaPlex is a real-time, full-duplex speech-to-speech conversational AI model from NVIDIA. It lets you control the assistant's persona through text role prompts and condition its voice on an audio sample, producing natural low-latency spoken conversations.
+3dsh-voice-ai-girlfriend
beiyege-01/dsh-voice-ai-girlfriend · ★ 1.1KIt is a voice AI girlfriend plugin for DeepSeek Harness (DSH): you talk to it through your microphone, it listens, chats back, and reads its replies aloud in a cloned voice while an animated digital-human avatar window reacts on screen.
The all-time top
5 projects · by stars
- 12.1K
#1 · NVIDIA/personaplexpersonaplex
PersonaPlex is a real-time, full-duplex speech-to-speech conversational AI model from NVIDIA. It lets you control the assistant's persona through text role prompts and condition its voice on an audio sample, producing natural low-latency spoken conversations.
- 1.3K
#2 · soniqo/speech-swiftspeech-swift
Speech Swift is a collection of on-device AI speech models for Apple Silicon — speech recognition (ASR), text-to-speech, speech-to-speech, voice activity detection, and speaker diarization — built as Swift packages powered by MLX and CoreML.
- 1.1K
#3 · beiyege-01/dsh-voice-ai-girlfrienddsh-voice-ai-girlfriend
It is a voice AI girlfriend plugin for DeepSeek Harness (DSH): you talk to it through your microphone, it listens, chats back, and reads its replies aloud in a cloned voice while an animated digital-human avatar window reacts on screen.
- 4

ai-native-callcenter
677rasonyang/ai-native-callcenter · GoAn open-source, self-hosted AI call center: calls land on FreeSWITCH and are answered by a speech-to-speech voice model by default, with a flow deciding when to hand the caller to a real human agent in a queue. It ships as one Go binary with a REST+SSE API, OpenAPI spec, PostgreSQL and FreeSWITCH underneath.
- 5

sophon-realtime
436adventists-ai/sophon-realtime · PythonAn open-source family of full-duplex voice models: it listens while you talk, hears your tone and answers with a voice that can laugh, sigh and breathe.
Related topics
Often tagged together