GGUF
conversational
Raja Muhammad Ali Asghar commited on
Commit
f2bcd70
Β·
verified Β·
1 Parent(s): 7c16335

Document VectorVision data path and official DOI

Browse files
Files changed (1) hide show
  1. README.md +31 -25
README.md CHANGED
@@ -1,25 +1,31 @@
1
- ---
2
- license: other
3
- license_name: bsl-1.1
4
- license_link: LICENSE
5
- models:
6
- - Qwen/Qwen2.5-0.5B-Instruct-GGUF
7
- ---
8
-
9
- # NEXIS-4 Base Model
10
-
11
- This repository holds the text-planning base and original RMAA materials for the NEXIS-4 music project.
12
-
13
- ## Current base model
14
-
15
- - `qwen2.5-0.5b-instruct-q2_k.gguf` β€” Qwen2.5-0.5B-Instruct Q2_K for original lyrics, song sections, and composition planning. It is text-only and does not generate sung audio. Upstream license: Apache-2.0; see `LICENSE-APACHE-2.0.txt`.
16
-
17
- ## NEXIS-4 voicebank policy
18
-
19
- All four NEXIS-4 singer identities (SOLA, VEXA, LUMEN, KAIYA) are to be crafted from synthetic-only training data. **No human voice recordings, third-party speaker presets, or pretrained human singer identities are included or used as NEXIS-4 voices.** The voicebanks and singing renderer are still to be built.
20
-
21
- Target song structure: Intro β†’ Verse 1 β†’ Chorus 1 β†’ Rap Verse β†’ Chorus 2 β†’ Verse 3 β†’ Chorus 3 β†’ Bridge β†’ Outro. Guitar is excluded from the default arrangement.
22
-
23
- ## Licensing
24
-
25
- `LICENSE` (BSL 1.1) covers only RMAA-original NEXIS-4 materials. The Qwen GGUF remains under its upstream Apache-2.0 license. See `MODEL_LICENSES.md`.
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ license_name: bsl-1.1
4
+ license_link: LICENSE
5
+ models:
6
+ - Qwen/Qwen2.5-0.5B-Instruct-GGUF
7
+ ---
8
+
9
+ # NEXIS-4 Base Model
10
+
11
+ This repository holds the text-planning base and original RMAA materials for the NEXIS-4 music project.
12
+
13
+ ## Current base model
14
+
15
+ - `qwen2.5-0.5b-instruct-q2_k.gguf` β€” Qwen2.5-0.5B-Instruct Q2_K for original lyrics, song sections, and composition planning. It is text-only and does not generate sung audio. Upstream license: Apache-2.0; see `LICENSE-APACHE-2.0.txt`.
16
+
17
+ ## NEXIS-4 voicebank policy
18
+
19
+ All four NEXIS-4 singer identities (SOLA, VEXA, LUMEN, KAIYA) are to be crafted from synthetic-only training data. **No human voice recordings, third-party speaker presets, or pretrained human singer identities are included or used as NEXIS-4 voices.** The voicebanks and singing renderer are still to be built.
20
+
21
+ Target song structure: Intro β†’ Verse 1 β†’ Chorus 1 β†’ Rap Verse β†’ Chorus 2 β†’ Verse 3 β†’ Chorus 3 β†’ Bridge β†’ Outro. Guitar is excluded from the default arrangement.
22
+
23
+ ## Licensing
24
+
25
+ `LICENSE` (BSL 1.1) covers only RMAA-original NEXIS-4 materials. The Qwen GGUF remains under its upstream Apache-2.0 license. See `MODEL_LICENSES.md`.
26
+
27
+ ## VectorVision data
28
+
29
+ The vector corpus is stored in [`vector/parquet`](https://huggingface.co/rmaapk/nexis4/tree/main/vector/parquet).
30
+
31
+ Official DOI: [10.57967/hf/11502](https://doi.org/10.57967/hf/11502).