Add pipeline tag
#2
by nielsr HF Staff - opened
README.md
CHANGED
|
@@ -1,5 +1,6 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
|
|
|
| 3 |
---
|
| 4 |
|
| 5 |
<h1>FireRedAudio</h1>
|
|
@@ -128,7 +129,8 @@ print(res.answer)
|
|
| 128 |
# ---- 2) Audio understanding (with optional chain-of-thought) --------------
|
| 129 |
res = engine.understand(
|
| 130 |
"assets/examples/assets_mmau_test.wav",
|
| 131 |
-
"What illness did Second speaker's friend suffer from?
|
|
|
|
| 132 |
task="understand", enable_thinking=True, max_new_tokens=10240,
|
| 133 |
)
|
| 134 |
print("CoT:")
|
|
@@ -177,7 +179,8 @@ uv run inference.py --task asr --model pretrained_models/FireRedAudio --audio as
|
|
| 177 |
# audio understanding and QA; several --audio for e.g. speaker verification,
|
| 178 |
# --enable-thinking to let the model reason first
|
| 179 |
uv run inference.py --task understand --model pretrained_models/FireRedAudio --audio assets/examples/assets_mmau_test.wav \
|
| 180 |
-
--prompt "What illness did Second speaker's friend suffer from?
|
|
|
|
| 181 |
|
| 182 |
|
| 183 |
# ICL voice cloning from a reference audio and its transcript
|
|
@@ -487,4 +490,4 @@ uv run inference.py --task voice_design --model pretrained_models/FireRedAudio -
|
|
| 487 |
|
| 488 |
## License
|
| 489 |
|
| 490 |
-
Released under the [Apache-2.0](LICENSE) license.
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
pipeline_tag: any-to-any
|
| 4 |
---
|
| 5 |
|
| 6 |
<h1>FireRedAudio</h1>
|
|
|
|
| 129 |
# ---- 2) Audio understanding (with optional chain-of-thought) --------------
|
| 130 |
res = engine.understand(
|
| 131 |
"assets/examples/assets_mmau_test.wav",
|
| 132 |
+
"What illness did Second speaker's friend suffer from?
|
| 133 |
+
(A) Progressive arthritis (B) Progressive cancer (C) Acute pneumonia (D) Chronic heart disease",
|
| 134 |
task="understand", enable_thinking=True, max_new_tokens=10240,
|
| 135 |
)
|
| 136 |
print("CoT:")
|
|
|
|
| 179 |
# audio understanding and QA; several --audio for e.g. speaker verification,
|
| 180 |
# --enable-thinking to let the model reason first
|
| 181 |
uv run inference.py --task understand --model pretrained_models/FireRedAudio --audio assets/examples/assets_mmau_test.wav \
|
| 182 |
+
--prompt "What illness did Second speaker's friend suffer from?
|
| 183 |
+
(A) Progressive arthritis (B) Progressive cancer (C) Acute pneumonia (D) Chronic heart disease" --enable-thinking --max-new-tokens 4096
|
| 184 |
|
| 185 |
|
| 186 |
# ICL voice cloning from a reference audio and its transcript
|
|
|
|
| 490 |
|
| 491 |
## License
|
| 492 |
|
| 493 |
+
Released under the [Apache-2.0](LICENSE) license.
|