fixie-ai ultravox issues

fixie-ai / ultravox

MIT License

612 stars 23 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Add adapter for HiSanta data

#47 mdepinet opened 3 days ago
3
Add streaming support, using HF TextStreamer

#46 juberti closed 4 days ago
0
Establish evaluation benchmark for conversational output

#45 juberti opened 6 days ago
0
Initial demo of speech output (tokenized)

#44 juberti opened 6 days ago
0
Create tech report for 0.2 including eval on BoolQ dataset

#43 juberti opened 6 days ago
0
Training run against Llama 3 70B

#42 juberti opened 6 days ago
0
Demo of emotion detection using BLSP-EMO approach

#41 juberti opened 6 days ago
0
Re-TTS AnyInstruct data using higher quality TTS

#40 juberti opened 6 days ago
0
Multilingual evaluation using multilingual conversational dataset

#39 juberti opened 6 days ago
0
Get Ultravox running with an initial hosting partner

#38 juberti opened 6 days ago
0
Land inference example in vLLM server

#37 juberti opened 6 days ago
0
Minor fixes to ds_tool and infer_tool

#36 juberti closed 4 days ago
0
SODA Dataset for Training

#35 farzadab opened 1 week ago
1
Don't fail basic inference due to missing OAI key

#34 juberti closed 1 week ago
0
Add 8-bit and 4-bit quant support

#33 juberti opened 2 weeks ago
2
Extending `ds_tool` for SODA conversational dataset

#32 farzadab closed 4 days ago
0
Add Eleven and Fireworks support to ds_tool

#31 juberti closed 2 weeks ago
0
BoolQ for Training and Eval

#30 farzadab closed 1 week ago
0
Allow for toggling training and eval on/off

#29 farzadab closed 2 weeks ago
0
Add @file syntax to ds_tool

#28 juberti closed 2 weeks ago
0
entails -> provides a rationale for

#27 farzadab closed 2 weeks ago
0
datasets: Handle converting `int16` audio data in `VoiceSample`.

#26 shaper closed 2 weeks ago
3
Create requirements.txt

#25 aksj98 closed 2 weeks ago
1
Switch from Pip to Poetry

#24 juberti closed 2 weeks ago
0
Remove mlflow dependency

#23 juberti closed 2 weeks ago
0
model: Add initial `HubertWithKmeans` module and extract tool.

#22 shaper closed 1 week ago
9
Don't run tests on docs changes

#21 juberti closed 2 weeks ago
0
Smaller model?!

#20 chukfinley opened 2 weeks ago
2
Add logging code

#19 juberti closed 2 weeks ago
1
Migrate pip setup to Poetry

#18 juberti closed 2 weeks ago
0
Update Hugging Face link

#17 simonw closed 2 weeks ago
1
Local tokenizer and processor for more consistent CI

#16 farzadab closed 2 weeks ago
2
Tool for uploading to HF Hub

#15 farzadab closed 2 weeks ago
0
Tool for adding new synthetic columns

#14 farzadab closed 2 weeks ago
2
Fixes the default HF model name

#13 cezarc1 closed 2 weeks ago
1
Add tts_tool, for converting a HF dataset to audio

#12 juberti closed 3 weeks ago
0
Add basic Jupyter notebook for inference

#11 juberti opened 3 weeks ago
0
Experiment with speech/text interleaving

#10 juberti opened 3 weeks ago
3
Perform steering finetune

#9 juberti opened 3 weeks ago
0
Evaluate Ultravox performance when quantized

#8 juberti opened 3 weeks ago
0
Add trivia_qa evaluation dataset

#7 juberti opened 3 weeks ago
0
Optionally include "passage" in BoolQ samples

#6 juberti closed 3 weeks ago
0
Whisper encoder + No 30 second padding

#5 farzadab closed 3 weeks ago
0
Audio Encoder to bfloat16

#4 farzadab closed 3 weeks ago
0
Any samples for local inference?

#3 Wyswyss closed 3 weeks ago
1
Improved Evaluations

#2 farzadab closed 4 weeks ago
1
Remove legacy directory

#1 farzadab closed 1 month ago
0