qwen-asr2api and qwen-tts2api
These are complementary tools that together create a bidirectional speech pipeline: ASR converts speech input to text while TTS converts text output to speech, enabling full voice interaction workflows.
About qwen-asr2api
aahl/qwen-asr2api
🎤 Qwen 3 ASR to OpenAI API, 免费STT语音识别模型
Wraps Alibaba's Qwen 3 ASR model as an OpenAI-compatible `/v1/audio/transcriptions` endpoint, enabling drop-in replacement for speech-to-text pipelines. Supports inverse text normalization (ITN) for number/punctuation normalization and deploys via Docker or as a Home Assistant add-on for local voice processing. The API accepts multipart audio uploads and returns transcriptions matching OpenAI's STT interface specification.
About qwen-tts2api
aahl/qwen-tts2api
🗣️ Qwen TTS to OpenAI Speech API
Implements an OpenAI-compatible Speech API wrapper around Alibaba's Qwen TTS model, supporting multiple voice profiles and audio output formats. Deployable via Docker or as a Home Assistant add-on, it provides a drop-in replacement for OpenAI's speech endpoint at `/v1/audio/speech`. Integrates seamlessly with Home Assistant's AI Conversation framework and any client expecting the standard OpenAI TTS interface.
Scores updated daily from GitHub, PyPI, and npm data. How scores work