Pinned issues
Issues
- 6
task: Ichigo LLM v0.5 Training
#122 opened by hahuyhoang411 - 0
milestone: Ichigo v0.5 Multi-lingual
#116 opened by hahuyhoang411 - 19
training run: tuning ichigo-quantizer
#144 opened by hahuyhoang411 - 13
planning: Train our own Quantizer for multilingual speech to work with Whisper Encoder
#146 opened by hahuyhoang411 - 1
- 18
task: Train and test text2semantic under decoder only framework for ichigo v0.5
#145 opened by tikikun - 1
- 2
chore: Whisper model demo
#147 opened by tikikun - 3
- 0
RQBottleneckTransformer.load_model("whisper-vq-stoks-v3-7lang-fixed.model") gives error
#149 opened by MatthewCYM - 1
feat: More multilingual audio training data
#143 opened by hahuyhoang411 - 0
idea: push Ichago inference to upstream llama.cpp
#142 opened by bennmann - 4
training run: Lora Ichigo Qwen2.5 32B
#134 opened by bachvudinh - 3
planning: Ichigo Transcription
#90 opened by dan-menlo - 6
research: Flow Matching for synthetic data generation
#140 opened by tikikun - 1
planning: Ichigo Encoder
#92 opened by dan-menlo - 2
planning: Ichigo Decoder
#111 opened by PodsAreAllYouNeed - 7
bug: [DESCRIPTION] SFT trainer does not use GPU
#104 opened by AvivSham - 9
epic: Reviews of Ichigo paper
#99 opened by tikikun - 6
planning: Ichigo v0.4 Speech Enhancements
#93 opened by dan-menlo - 4
epic: Introducing naturalness for interuption
#100 opened by tikikun - 12
feat: fast open source tts for ichigo
#94 opened by PodsAreAllYouNeed - 1
data: Singlish Dataset
#135 opened by dan-menlo - 6
- 0
training run: Qwen 0.5B as Text to semantic model
#137 opened by bachvudinh - 1
Research: Downloading National Speech Corpus
#138 opened by bachvudinh - 0
task: MLX quantizer to support productionize ichigo
#141 opened by tikikun - 2
- 1
worklog: Ichigo v0.5 development
#136 opened by hahuyhoang411 - 1
idea: Training Ichigo on Structured Output
#126 opened by hahuyhoang411 - 0
data: Dataset Collection & Preprocessing
#119 opened by hahuyhoang411 - 6
Some Question about the technology report.
#132 opened by handsomelys - 2
bug: The demo is frozen sometimes on multi-turn
#105 opened by tikikun - 1
License?
#86 opened by ignoramous - 1
Utilizing Ichigo Without Self-Training
#107 opened by ruidazeng - 4
feat: Ichigo from semantics to sound tokens
#108 opened by tikikun - 3
- 0
- 7
idea: Precision scaling research
#127 opened by hahuyhoang411 - 0
bug: Ichigo Notebook v0.4 not working
#128 opened by tuanlda78202 - 1
How can the voice be changed?
#85 opened by cshipley - 4
epic: Replace optimizers or code new optimizers?
#103 opened by tikikun - 3
Can it run on Apple Silicon by any chance?
#101 opened by ahmetkca - 1
idea: official docker image / container builds
#95 opened by sammcj - 4
planning: Ichigo VAD
#91 opened by dan-menlo - 4
feat: Modify Torchtune to train Ichigo Qwen
#125 opened by hahuyhoang411 - 2
run: ichigo 0.4
#102 opened by tikikun - 2
Reproducing the OpenHermes and Alpaca Audio Evals
#98 opened by Helw150 - 0
- 0
epic: [DESCRIPTION]
#117 opened by bachvudinh