Conversation
## NLP Improvements - Port improved Japanese normalizer with unit/date/currency handling - Add katakana_map.py for katakana normalization - Port mora_list.py with pyopenjtalk-plus morae (15+ new morae) - Port robust adjust_word2ph() fallback in g2p.py - Port mora validation rules in user_dict/word_model.py - Add English short_form_dict for abbreviations (89 entries) - Simplify pyopenjtalk_worker with auto-reconnect logic ## Inference Optimizations - Add infer_stream for streaming TTS output - Implement model_dtype parameter for FP16/BF16 inference - Chunk-based decoder processing with overlap-add ## API & UI - Add /voice/stream streaming endpoint to server_fastapi.py - Add streaming UI to Gradio with chunk/overlap config ## Config Simplification - Simplify config.yml structure - Remove redundant configuration classes ## Testing & Documentation - Add benchmark_dtype.py for precision comparison - Add streaming_benchmark.py for streaming tests - Add test_normalizer.py for Japanese normalizer tests - Add FORK_COMPARISON.md documenting differences with fork 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Remove ZH/EN language support (JP-only) - Add "v" phoneme to JP_SYMBOLS for ヴ sounds - Refactor WavLM to use HuggingFace repo directly - Add convert_model_v3.py for v2->v3 model conversion - Update initialize.py to download v3 pretrained models - Support multiple audio formats in transcribe/resample - Remove local slm/wavlm-base-plus files - Various code cleanups and documentation updates 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
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
1 participant
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.
requirements.txtでのインストールをやめて、uvでpyproject.tomlのdev dependenciesとしてすべて扱うfaster_whisperを廃止、TransforemrsライブラリのWhisperを常に使うtorch2.9系にするpyopenjtalk-plusを使うTODO