Add Qwen3-ASR-0.6B WASM transcriber as alternative to Moonshine - #8
Open
ozankaraali wants to merge 1 commit into
Open
Add Qwen3-ASR-0.6B WASM transcriber as alternative to Moonshine#8ozankaraali wants to merge 1 commit into
ozankaraali wants to merge 1 commit into
Conversation
Compile qwen3-asr.cpp (GGML-based C++ implementation) to WebAssembly via Emscripten, producing a 865KB WASM binary for browser inference. New files: - lib/qwen3-asr-wasm.wasm: GGML inference engine compiled to WASM - lib/qwen3-asr-wasm.js: Emscripten JS glue code - lib/transcriber-qwen3.js: JS integration with same LimitrTranscriber API as the Moonshine transcriber (drop-in replacement) Features: - IndexedDB caching of the GGUF model file after first download - Same capture worklet + latest-wins audio pipeline as Moonshine - WASM SIMD 128-bit for optimized tensor operations - 52-language multilingual ASR support The GGUF model (~1.3GB Q8_0) must be generated from the official Qwen/Qwen3-ASR-0.6B weights using qwen3-asr.cpp's convert script and hosted externally. See comments in transcriber-qwen3.js. https://claude.ai/code/session_01GxACrvA8d54L4W9ryYrfkf
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Compile qwen3-asr.cpp (GGML-based C++ implementation) to WebAssembly
via Emscripten, producing a 865KB WASM binary for browser inference.
New files:
API as the Moonshine transcriber (drop-in replacement)
Features:
The GGUF model (~1.3GB Q8_0) must be generated from the official
Qwen/Qwen3-ASR-0.6B weights using qwen3-asr.cpp's convert script
and hosted externally. See comments in transcriber-qwen3.js.
https://claude.ai/code/session_01GxACrvA8d54L4W9ryYrfkf