kitlangton / kitlangton/Hex

Feature Request: Consider SenseVoice for faster voice-to-text

Open
#233 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Swift
Stars
2.9k
Forks
226
PR merge metrics
No merged PRs in 30d

Description

## Feature Request

Hex converts voice to words. Suggesting [SenseVoice](https://github.com/FunAudioLLM/SenseVoice) as a model option for faster transcription.

### Why SenseVoice?

- **Non-autoregressive** — instant results
- **5x faster than Whisper** — lower latency between speaking and seeing text
- **234M params** — lightweight
- **50+ languages** — multilingual
- **macOS ready** — via [Sherpa-ONNX](https://github.com/k2-fsa/sherpa-onnx) Swift bindings with Apple Silicon optimization

- FunASR: https://github.com/modelscope/FunASR (16.6K stars)
- SenseVoice: https://github.com/FunAudioLLM/SenseVoice (8.3K stars)

Contributor guide

No contributing guide indexed for this repository

Research direction

Start by inspecting the existing Swift transcription and Whisper integration, then review the linked SenseVoice and Sherpa-ONNX projects and their macOS Swift bindings. Done means SenseVoice is available as a model option and provides faster voice-to-text transcription on Apple Silicon.

Written by the indexing model from the issue text.

Assessment

Tech stack
macos, swift
Domain
audio-video-rtc, desktop
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.