Brandon Thomas
f096ae07bf
Changing proto schema. Degradation in quality.
...
I had something really close. That might not have found it into a
commit between now and the last commit :(
2019-10-21 02:58:05 -04:00
Brandon Thomas
af20a57e8a
THIS EVALUATES THE MODEL IN REALTIME +3-6sec delay
...
Still doesn't sound right. Probably a sample rate issue.
2019-10-20 03:10:21 -04:00
Brandon Thomas
d63007ac9f
Experimenting - trying to figure out why audio sounds so bad
2019-10-20 02:50:18 -04:00
Brandon Thomas
7e33d9c7ca
Add my demo_server.py implementation from CycleGAN
2019-10-20 02:18:51 -04:00
Brandon Thomas
b38c96ae2f
Protobufs UP and DOWN. Sounds great!
2019-10-20 02:04:41 -04:00
Brandon Thomas
05b703068a
THIS SOUNDS FANTASTIC (no transform yet, 1:1 audio)
2019-10-20 01:51:55 -04:00
Brandon Thomas
6df1e8b072
Two way works over protobuf! Woo! This will be so much better...
2019-10-20 01:36:12 -04:00
Brandon Thomas
dc9983e45b
Protos in python
2019-10-20 01:17:25 -04:00
Brandon Thomas
45a8dc446e
Better proto pipeline
2019-10-20 00:48:18 -04:00
Brandon Thomas
1ae8aafd5a
Add protos. Temporarily commented out TensorFlow pieces
2019-10-20 00:38:04 -04:00
Brandon Thomas
47022cdeeb
Copy relevant CycleGAN code here.
...
master @ 1ca8b078c12d915bca1f57e822859fa20bd439c8
Date: Tue Oct 15 10:50:12 2019 -0400
2019-10-20 00:11:53 -04:00
Brandon Thomas
7c3f3255b0
STUPID VEC POP WAS THE ISSUE
2019-10-15 02:59:16 -04:00
Brandon Thomas
4b956e594f
get rid of unneeded loop
2019-10-15 02:45:18 -04:00
Brandon Thomas
86900188d8
super direct round trip
2019-10-15 01:49:38 -04:00
Brandon Thomas
708a728b85
simple audio roundtrip
2019-10-15 01:41:24 -04:00
Brandon Thomas
de07378bcb
cleanup!
2019-10-15 01:20:37 -04:00
Brandon Thomas
7916f7c307
this inquires about processed buffer length
2019-10-15 00:55:12 -04:00
Brandon Thomas
a6ac91331e
whellp, there's the data round trip. This is bad
2019-10-15 00:35:34 -04:00
Brandon Thomas
ff0a7d7922
sending successfully
2019-10-14 20:00:21 -04:00
Brandon Thomas
6464e6344d
fix formatting
2019-10-14 19:18:21 -04:00
Brandon Thomas
9f11012252
add CPAL
2019-10-14 19:14:47 -04:00
Brandon Thomas
8f611958a3
sending more data is better
2019-10-13 22:20:10 -04:00
Brandon Thomas
6d928fa7ec
cleanup speaker/mic
2019-10-13 21:49:23 -04:00
Brandon Thomas
64e4018640
This works with large byte buffers! Sidecar reconstitutes the audio
2019-10-13 21:42:32 -04:00
Brandon Thomas
592bebe823
drained send
2019-10-13 21:35:43 -04:00
Brandon Thomas
089aab7930
move model stuff to new file
2019-10-13 21:23:11 -04:00
Brandon Thomas
58cfbef65e
sending audio, but still very broken
2019-10-13 20:51:18 -04:00
Brandon Thomas
e88cc5ff24
ZeroMQ IPC (first steps)
2019-10-13 00:41:46 -04:00
Brandon Thomas
08d7fc6797
zeromq
2019-10-12 23:21:37 -04:00
Brandon Thomas
41e9e45fe7
start work on synthesis
2019-10-11 01:58:39 -04:00
Brandon Thomas
af2f5ba68e
Modify readme
2019-10-11 00:21:54 -04:00
Brandon Thomas
cd6b5a37fd
change how dep override is declared
2019-09-30 01:43:44 -04:00
Brandon Thomas
a66a084a84
same dimensions for input!
2019-09-23 20:02:00 -04:00
Brandon Thomas
3cd90ded8b
more OO
2019-09-23 01:42:25 -04:00
Brandon Thomas
5cc533061e
This returns a tensor...
2019-09-23 01:25:47 -04:00
Brandon Thomas
2377494231
Getting closer...
2019-09-23 01:23:30 -04:00
Brandon Thomas
b9bb56085e
this works with the model
2019-09-23 01:18:55 -04:00
Brandon Thomas
2364f770f2
Now dimensional issues:
...
>>> Tensor dims: [2, 2, 2]
2019-09-23 01:13:51.546620: W
tensorflow/core/framework/op_kernel.cc:1502] OP_REQUIRES failed at
conv_ops_fused_impl.h:692 : Invalid argument: input depth must be evenly
divisible by filter depth: 2 vs 24
2019-09-23 01:13:51.546627: W
tensorflow/core/framework/op_kernel.cc:1502] OP_REQUIRES failed at
conv_ops_fused_impl.h:692 : Invalid argument: input depth must be evenly
divisible by filter depth: 2 vs 24
thread 'main' panicked at 'Run success: {inner:0x5581658b56d0,
InvalidArgument: input depth must be evenly divisible by filter depth: 2
vs 24
[[{{node generator_A2B_3/h1_conv_gates/conv1d}}]]}',
src/libcore/result.rs:999:5
2019-09-23 01:14:34 -04:00
Brandon Thomas
2dd5894283
Found the names of the input and output tensors, now fixing args
...
Current runtime err:
```
2019-09-23 01:00:01.189765: W
tensorflow/core/framework/op_kernel.cc:1502] OP_REQUIRES failed at
transpose_op.cc:157 : Invalid argument: transpose expects a vector of
size 1. But input(1) is a vector of size 3
thread 'main' panicked at 'Run success: {inner:0x5589ea8b3860,
InvalidArgument: transpose expects a vector of size 1. But input(1) is a
vector of size 3
```
Getting there...!
2019-09-23 01:05:31 -04:00
Brandon Thomas
51b7fce129
Found the names of the input and output tensors, now fixing args
...
Current runtime err:
```
2019-09-23 01:00:01.189765: W
tensorflow/core/framework/op_kernel.cc:1502] OP_REQUIRES failed at
transpose_op.cc:157 : Invalid argument: transpose expects a vector of
size 1. But input(1) is a vector of size 3
thread 'main' panicked at 'Run success: {inner:0x5589ea8b3860,
InvalidArgument: transpose expects a vector of size 1. But input(1) is a
vector of size 3
```
Getting there...!
2019-09-23 01:00:55 -04:00
Brandon Thomas
3c715d9ef8
Merge branch 'master' of github.com:echelon/voder
2019-09-22 22:16:40 -04:00
Brandon Thomas
1201fc120a
this is a valid argument (used incorrectly)
2019-09-17 18:47:26 -04:00
Brandon Thomas
9f3f4f7d60
model loads!
2019-09-17 01:50:47 -04:00
Brandon Thomas
4f331f3f83
model loading works, but is from a wrong version of tensorflow
2019-09-16 11:18:16 -04:00
Brandon Thomas
2982a82999
feedback, adding tensorflow
2019-09-16 03:26:18 -04:00
Brandon Thomas
0483c2fe6e
comments and readme
2019-09-13 04:17:29 -04:00
Brandon Thomas
12440fe1fe
add old "main.rs" as "weird.rs", add debugging to beep
2019-09-13 02:14:15 -04:00
Brandon Thomas
9482fb2302
add wavy example code
2019-09-13 01:50:00 -04:00
Brandon Thomas
39bdce3380
Add more examples from cpal
2019-09-13 01:47:54 -04:00
Brandon Thomas
8479a64a6b
fixes in notes
2019-09-11 03:10:05 -04:00
Brandon Thomas