Adding temperature scaling on Joiner logits: (#789)

* Adding temperature scaling on Joiner logits:

- T hard-coded to 2.0
- so far best result NCE 0.122 (still not so high)
    - the BPE scores were rescaled with 0.2 (but then also incorrect words
      get high confidence, visually reasonable histograms are for 0.5 scale)
    - BPE->WORD score merging done by min(.) function
      (tried also prob-product, and also arithmetic, geometric, harmonic mean)

- without temperature scaling (i.e. scale 1.0), the best NCE was 0.032 (here product merging was best)

Results seem consistent with: https://arxiv.org/abs/2110.15222

Everything tuned on a very-small set of 100 sentences with 813 words and 10.2% WER, a Czech model.

I also experimented with blank posteriors mixed into the BPE confidences,
but no NCE improvement found, so not pushing that.

Temperature scling added also to the Greedy search confidences.

* making `temperature_scale` configurable from outside

sherpa-onnx/csrc/online-transducer-greedy-search-decoder.h

@@ -15,8 +15,13 @@ namespace sherpa_onnx {
 class OnlineTransducerGreedySearchDecoder : public OnlineTransducerDecoder {
  public:
   OnlineTransducerGreedySearchDecoder(OnlineTransducerModel *model,
-                                      int32_t unk_id, float blank_penalty)
-      : model_(model), unk_id_(unk_id), blank_penalty_(blank_penalty) {}
+                                      int32_t unk_id,
+                                      float blank_penalty,
+                                      float temperature_scale)
+      : model_(model),
+      unk_id_(unk_id),
+      blank_penalty_(blank_penalty),
+      temperature_scale_(temperature_scale) {}
 
   OnlineTransducerDecoderResult GetEmptyResult() const override;
 
@@ -29,6 +34,7 @@ class OnlineTransducerGreedySearchDecoder : public OnlineTransducerDecoder {
   OnlineTransducerModel *model_;  // Not owned
   int32_t unk_id_;
   float blank_penalty_;
+  float temperature_scale_;
 };
 
 }  // namespace sherpa_onnx