Total runs:
1.5M
Run Growth:
271.0K
Growth Rate:
18.35%
Sponsored by
APIMart
litert-community / TIGER-DnR-LiteRT
Model's Last Updated: September 08 2026
audio-to-audio
Introduction of TIGER-DnR-LiteRT
Model Details of TIGER-DnR-LiteRT
TIGER-DnR — LiteRT (CompiledModel GPU) cinematic sound separation
Mixture input → separated Dialogue / Sound effects / Music stems, from the on-device model (waveforms).
TIGER
(ICASSP 2025) DnR
cinematic sound separation
re-authored to
GPU-native
LiteRT
.tflite
: split a clip into
Dialogue / Sound effects / Music
stems fully on-device.
Three sibling band-split TIGER graphs (~1.41 M params each, trained on the openly-built
DnR
dataset by the upstream authors) each process a
12.06 s,
44.1 kHz mono
chunk; per DnR convention each graph contributes one stem (dialog = source 2 of
tiger_dialog
, effect = source 1 of
tiger_effect
, music = source 0 of
tiger_music
).
Verified on a Pixel 8a (Tensor G3): full LITERT_CL residency — 23 974 / 23 974 nodes, 1 partition per graph, GPU output vs PyTorch waveform corr 0.99987 , ~4.5 s per chunk per graph. FP16, 16.1 MB per graph.
Files
| file | stem it contributes |
|---|---|
tiger_dialog_fp16.tflite
|
Dialogue (output source index 2) |
tiger_effect_fp16.tflite
|
Sound effects (output source index 1) |
tiger_music_fp16.tflite
|
Music (output source index 0) |
I/O
-
Input
[1, 534016]float32 — a 531 968-sample (12.06 s @ 44.1 kHz) mono chunk, reflect-padded by 1024 samples on both sides by the caller (torch.stft(center=True)equivalent). The STFT (win 2048 / hop 512, periodic Hann) runs inside the graph as a windowed-DFTConv1d. -
Outputs
[1, 3, 1025, 1040]× 2 (real, imag) — separated one-sided complex spectrograms for the 3 sources. Host runs iSTFT (win 2048 / hop 512, Hann, trim 1024 center pad) and overlap-adds chunks (reference inference uses a 12 s window with 4–10 s hop, averaging overlaps).
Minimal usage (Python, desktop parity check)
import numpy as np, soundfile as sf, torch
from ai_edge_litert.interpreter import Interpreter
SR, WIN, HOP, T = 44100, 2048, 512, 1040
S = (T - 1) * HOP # 531968 samples = 12.06 s chunk
wav, _ = sf.read("mix.wav", dtype="float32") # mono 44.1 kHz
x = np.zeros(S, np.float32); n = min(len(wav), S); x[:n] = wav[:n]
x = np.concatenate([x[WIN//2:0:-1], x, x[-2:-WIN//2-2:-1]]) # reflect pad (host side)
it = Interpreter(model_path="tiger_dialog_fp16.tflite"); it.allocate_tensors()
it.set_tensor(it.get_input_details()[0]["index"], x[None]); it.invoke()
real, imag = (it.get_tensor(o["index"]) for o in
sorted(it.get_output_details(), key=lambda o: o["index"])) # [1,3,1025,1040] x2
spec = torch.complex(torch.tensor(real), torch.tensor(imag))[0, 2] # source 2 = dialogue
dialog = torch.istft(spec, n_fft=WIN, hop_length=HOP,
window=torch.hann_window(WIN), length=S)
sf.write("dialog.wav", dialog.numpy(), SR)
# effects: tiger_effect_fp16.tflite source 1 · music: tiger_music_fp16.tflite source 0
Kotlin (Android, LiteRT CompiledModel GPU)
// implementation("com.google.ai.edge.litert:litert:2.1.5")
val model = CompiledModel.create(File(ctx.filesDir, "tiger_dialog_fp16.tflite").absolutePath,
CompiledModel.Options(Accelerator.GPU), null)
val inBuf = model.createInputBuffers()
val outBuf = model.createOutputBuffers()
// 12.06 s mono 44.1 kHz chunk (531968 samples), reflect-padded by 1024 on both sides -> 534016
inBuf[0].writeFloat(paddedChunk)
model.run(inBuf, outBuf)
val real = outBuf[0].readFloat() // [3 * 1025 * 1040], source-major: src*1025*1040 + f*1040 + t
val imag = outBuf[1].readFloat()
// dialogue = source index 2; host iSTFT (win 2048, hop 512, periodic Hann, trim 1024 center pad)
// + overlap-add across chunks — see Istft.kt in the audio_source_separation LiteRT sample.
Conversion (numerically-equivalent re-authoring)
torch.stft
→ in-graph DFT-as-
Conv1d
(host reflect-pad);
torch.istft
→ host; folded-batch
Conv1d
(
(B·T, N, band)
) → 4D
(1,k)
-
Conv2d
; per-sample
GlobLN
→ per-position chained-single-axis-mean
SafeNorm; chunk length chosen so T=1040 is divisible by 16 → adaptive pools become uniform
AVERAGE_POOL_2D
and nearest resizes exact integer repeats; the non-uniform 57-band axis uses constant
averaging/one-hot
FULLY_CONNECTED
matrices; MHSA → per-head batch-1 3D matmuls with 1/√d folded into Q;
PReLU
→
relu(x) − w·relu(−x)
; the 6-D mask-head view → static channel slices. Two fp16-on-GPU fixes,
both exact-equivalent: norm eps raised to 1e-4 (1e-8/1e-5 underflow to 0 in fp16 → 0/0 NaN on silent
bands) and the mask head rewritten without dim-1 broadcast MULs.
Desktop parity: fp16 tflite vs PyTorch waveform corr 0.99991 (per-graph, real 12 s mixture).
Upstream
- Code: JusperLee/TIGER (MIT)
- Weights: JusperLee/TIGER-DnR (Apache-2.0)
Please cite the TIGER paper (Xu et al., ICASSP 2025, arXiv:2410.01469) when you use these models.
Runs of litert-community TIGER-DnR-LiteRT on huggingface.co
697
Total runs
-4
24-hour runs
0
3-day runs
-42
7-day runs
14
30-day runs
More Information About TIGER-DnR-LiteRT huggingface.co Model
Url of TIGER-DnR-LiteRT
TIGER-DnR-LiteRT huggingface.co Url
Provider of TIGER-DnR-LiteRT huggingface.co
Other API from litert-community
Updated:August 31 2026
Total runs:
462.7K
Run Growth:
99.8K
Growth Rate:
21.71%
Total runs:
93.6K
Run Growth:
93.6K
Growth Rate:
99.99%
Total runs:
91.4K
Run Growth:
19.9K
Growth Rate:
21.73%
Total runs:
61.3K
Run Growth:
-7.9K
Growth Rate:
-12.91%
Total runs:
47.3K
Run Growth:
3.8K
Growth Rate:
8.08%
Total runs:
23.2K
Run Growth:
-857
Growth Rate:
-3.69%
Total runs:
12.0K
Run Growth:
-365
Growth Rate:
-3.03%
Total runs:
11.8K
Run Growth:
-2.1K
Growth Rate:
-18.13%
Total runs:
9.8K
Run Growth:
-3.2K
Growth Rate:
-33.26%
Total runs:
9.7K
Run Growth:
1.8K
Growth Rate:
18.65%
Total runs:
9.1K
Run Growth:
9.1K
Growth Rate:
99.97%
Total runs:
7.9K
Run Growth:
529
Growth Rate:
6.70%
Total runs:
7.9K
Run Growth:
7.2K
Growth Rate:
90.97%
Total runs:
7.2K
Run Growth:
6.8K
Growth Rate:
93.53%
Total runs:
7.2K
Run Growth:
7.2K
Growth Rate:
99.93%
Total runs:
7.1K
Run Growth:
-451
Growth Rate:
-6.38%
Total runs:
6.7K
Run Growth:
0
Growth Rate:
0.00%
Total runs:
6.6K
Run Growth:
147
Growth Rate:
2.23%
Total runs:
5.9K
Run Growth:
2.6K
Growth Rate:
44.38%
Total runs:
4.5K
Run Growth:
686
Growth Rate:
15.31%
Total runs:
4.3K
Run Growth:
3.1K
Growth Rate:
71.59%
Total runs:
4.0K
Run Growth:
449
Growth Rate:
11.31%
Total runs:
3.8K
Run Growth:
1.2K
Growth Rate:
31.06%
Total runs:
3.7K
Run Growth:
180
Growth Rate:
4.84%
Total runs:
3.2K
Run Growth:
1.6K
Growth Rate:
48.21%
Total runs:
3.1K
Run Growth:
455
Growth Rate:
14.78%
Total runs:
2.8K
Run Growth:
2.8K
Growth Rate:
98.47%
Total runs:
2.8K
Run Growth:
-254
Growth Rate:
-9.18%
Total runs:
2.6K
Run Growth:
309
Growth Rate:
11.72%
Total runs:
2.5K
Run Growth:
0
Growth Rate:
0.00%
Total runs:
2.3K
Run Growth:
237
Growth Rate:
10.33%
Total runs:
2.2K
Run Growth:
1.5K
Growth Rate:
67.82%
Total runs:
2.1K
Run Growth:
2.1K
Growth Rate:
100.00%
Total runs:
2.1K
Run Growth:
-188
Growth Rate:
-9.20%
Total runs:
2.1K
Run Growth:
-96
Growth Rate:
-4.64%
Total runs:
2.0K
Run Growth:
891
Growth Rate:
43.78%
Total runs:
1.7K
Run Growth:
-285
Growth Rate:
-17.23%
Total runs:
1.6K
Run Growth:
-287
Growth Rate:
-18.16%
Total runs:
1.5K
Run Growth:
-300
Growth Rate:
-19.83%
Total runs:
1.5K
Run Growth:
0
Growth Rate:
0.00%
Total runs:
1.5K
Run Growth:
495
Growth Rate:
33.51%
Total runs:
1.5K
Run Growth:
-64
Growth Rate:
-4.19%
Total runs:
1.3K
Run Growth:
984
Growth Rate:
73.21%
Total runs:
1.3K
Run Growth:
-107
Growth Rate:
-8.04%
Total runs:
1.3K
Run Growth:
-952
Growth Rate:
-74.43%
Total runs:
1.3K
Run Growth:
982
Growth Rate:
78.37%
Total runs:
1.3K
Run Growth:
-486
Growth Rate:
-38.79%
Total runs:
1.2K
Run Growth:
551
Growth Rate:
44.62%
Total runs:
1.2K
Run Growth:
987
Growth Rate:
81.23%
Total runs:
1.1K
Run Growth:
-65
Growth Rate:
-5.80%
Total runs:
984
Run Growth:
0
Growth Rate:
0.00%
Total runs:
968
Run Growth:
40
Growth Rate:
4.13%
Total runs:
956
Run Growth:
935
Growth Rate:
97.80%
Total runs:
954
Run Growth:
398
Growth Rate:
41.72%
Total runs:
931
Run Growth:
-139
Growth Rate:
-14.93%
Total runs:
798
Run Growth:
733
Growth Rate:
91.85%
Total runs:
724
Run Growth:
724
Growth Rate:
100.00%
Total runs:
716
Run Growth:
419
Growth Rate:
58.52%
Total runs:
711
Run Growth:
220
Growth Rate:
30.94%
Total runs:
662
Run Growth:
-2
Growth Rate:
-0.30%
Total runs:
645
Run Growth:
88
Growth Rate:
13.64%
Total runs:
628
Run Growth:
292
Growth Rate:
46.50%
Total runs:
613
Run Growth:
455
Growth Rate:
74.23%
Total runs:
560
Run Growth:
-656
Growth Rate:
-117.14%
Total runs:
558
Run Growth:
144
Growth Rate:
25.81%
Total runs:
522
Run Growth:
490
Growth Rate:
93.87%
Total runs:
499
Run Growth:
337
Growth Rate:
67.54%
Total runs:
491
Run Growth:
116
Growth Rate:
23.63%
Total runs:
486
Run Growth:
486
Growth Rate:
100.00%
Total runs:
480
Run Growth:
238
Growth Rate:
49.58%
Total runs:
460
Run Growth:
230
Growth Rate:
50.00%
Total runs:
444
Run Growth:
116
Growth Rate:
26.13%
Total runs:
425
Run Growth:
425
Growth Rate:
100.00%
Total runs:
418
Run Growth:
105
Growth Rate:
25.12%
Total runs:
406
Run Growth:
210
Growth Rate:
51.72%
Total runs:
392
Run Growth:
172
Growth Rate:
43.88%
Total runs:
379
Run Growth:
-25
Growth Rate:
-6.60%
Total runs:
379
Run Growth:
13
Growth Rate:
3.43%
Total runs:
375
Run Growth:
58
Growth Rate:
15.47%
Total runs:
373
Run Growth:
-130
Growth Rate:
-34.85%
Total runs:
370
Run Growth:
193
Growth Rate:
52.16%
Total runs:
366
Run Growth:
200
Growth Rate:
54.64%
Total runs:
364
Run Growth:
246
Growth Rate:
67.58%
Total runs:
357
Run Growth:
357
Growth Rate:
100.00%
Total runs:
340
Run Growth:
36
Growth Rate:
10.59%
Total runs:
330
Run Growth:
201
Growth Rate:
60.91%
Total runs:
314
Run Growth:
130
Growth Rate:
41.40%
Total runs:
310
Run Growth:
16
Growth Rate:
5.16%
Total runs:
287
Run Growth:
-47
Growth Rate:
-16.38%
Total runs:
282
Run Growth:
145
Growth Rate:
51.42%
Total runs:
280
Run Growth:
-5
Growth Rate:
-1.79%
Total runs:
266
Run Growth:
-31
Growth Rate:
-11.65%
Total runs:
260
Run Growth:
170
Growth Rate:
65.38%
Total runs:
253
Run Growth:
99
Growth Rate:
39.13%
Total runs:
251
Run Growth:
2
Growth Rate:
0.80%
Total runs:
244
Run Growth:
128
Growth Rate:
52.46%
Total runs:
230
Run Growth:
7
Growth Rate:
3.06%