EnCodec

EnCodec is a Facebook/Meta developed AI means of compressing audio files using High Fidelity Neural Audio Compression. EnCodec is designed to provide codec compression at 6 kbps using their novel AI-powered compression technique. The test profile uses a lengthy JFK speech as the audio input for benchmarking and the performance measurement is measuring the time to encode the EnCodec file from WAV.


EnCodec 0.1.1

Target Bandwidth: 24 kbps

OpenBenchmarking.org metrics for this test profile configuration based on 260 public results since 2 November 2022 with the latest data as of 20 March 2024.

Below is an overview of the generalized performance for components where there is sufficient statistically significant data based upon user-uploaded results. It is important to keep in mind particularly in the Linux/open-source space there can be vastly different OS configurations, with this overview intended to offer just general guidance as to the performance expectations.

Component
Details
Percentile Rank
# Compatible Public Results
Seconds (Average)
Zen 4 [16 Cores / 32 Threads]
96th
8
22 +/- 1
Zen 4 [12 Cores / 24 Threads]
92nd
8
22 +/- 1
Raptor Lake [24 Cores / 32 Threads]
87th
13
23 +/- 1
Zen 4 [12 Cores / 24 Threads]
85th
5
23 +/- 1
Sapphire Rapids [60 Cores / 120 Threads]
80th
4
24 +/- 1
Zen 4 [8 Cores / 16 Threads]
80th
3
24 +/- 1
Mid-Tier
75th
> 26
Zen 4 [12 Cores / 24 Threads]
74th
7
26 +/- 3
Zen 4 [6 Cores / 12 Threads]
72nd
8
27 +/- 1
Raptor Lake [14 Cores / 20 Threads]
71st
3
28 +/- 1
Zen 4 [16 Cores / 32 Threads]
70th
11
28 +/- 3
Zen 4 [6 Cores / 12 Threads]
62nd
7
31 +/- 4
Rocket Lake [8 Cores / 16 Threads]
56th
7
32 +/- 1
Cascade Lake [18 Cores / 36 Threads]
52nd
4
33
Zen 3 [16 Cores / 32 Threads]
52nd
4
33 +/- 1
Sapphire Rapids [120 Cores / 240 Threads]
51st
4
33 +/- 1
Median
50th
33
Zen 3 [8 Cores / 16 Threads]
50th
3
33 +/- 1
Zen 3 [8 Cores / 16 Threads]
49th
3
34 +/- 1
Cascade Lake [16 Cores / 32 Threads]
44th
3
35 +/- 1
Raptor Lake [10 Cores / 16 Threads]
44th
4
35
Rocket Lake [6 Cores / 12 Threads]
41st
5
36 +/- 1
Haswell [16 Cores / 16 Threads]
41st
3
37 +/- 1
Zen 2 [32 Cores / 64 Threads]
39th
3
38
Alder Lake [10 Cores / 16 Threads]
36th
4
40
Zen 2 [8 Cores / 16 Threads]
35th
3
40
Zen 2 [24 Cores / 48 Threads]
31st
4
42
Zen 2 [64 Cores / 128 Threads]
29th
3
43
Zen 3 [8 Cores / 16 Threads]
28th
3
45
Kaby/Coffee/Whiskey Lake [6 Cores / 12 Threads]
26th
3
46 +/- 5
Low-Tier
25th
> 47
Comet Lake [8 Cores / 16 Threads]
24th
3
47
Zen 3 [128 Cores / 256 Threads]
20th
4
49 +/- 1
Cascade Lake [56 Cores / 112 Threads]
18th
4
50
Cascade Lake [48 Cores / 96 Threads]
16th
4
51 +/- 3
Tiger Lake [4 Cores / 8 Threads]
15th
3
53
Tiger Lake [4 Cores / 8 Threads]
15th
3
53 +/- 1
Zen 2 [12 Cores / 24 Threads]
12th
3
55
Zen 3 [8 Cores / 16 Threads]
11th
3
59
Skylake [8 Cores / 8 Threads]
10th
4
66 +/- 8
Zen [64 Cores / 128 Threads]
7th
4
80 +/- 2
Zen 2 [6 Cores / 12 Threads]
5th
4
86
Zen 2 [8 Cores / 8 Threads]
4th
3
90 +/- 2
Zen 2 [6 Cores / 6 Threads]
3rd
3
96 +/- 1
Zen 2 [8 Cores / 16 Threads]
2nd
3
119 +/- 2